下载文档原格式
Elsevier 投稿各种状态总结From /u2/85704/showart_2263346.html附录文章投稿状态Elsevier 投稿各种状态总结1. Submitted to Journal当上传结束后，显示的状态是Submitted to Journal，这个状态是自然形成的无需处理。
2. With editor如果在投稿的时候没有要求选择编辑，就先到主编那，主编会分派给别的编辑。
这当中就会3. Editor assigned4. Editor Declined Invitation如果编辑接手处理了就会邀请审稿人了。
5. Reviewer(s) invited如果审稿人接受那就会是以下状态：6. Under review这应该是一个漫长的等待。
当然前面各步骤也可能很慢的，要看编辑的处理情况。
如果被邀请审稿人不想审，就会decline，编辑会重新邀请别的审稿人。
7. required review completed审稿结束，等编辑处理。
8. Decision in Process到了这一步就快要有结果了，编辑开始考虑是给修改还是直接拒，当然也有可能直接接受的，9. Minor revision/Major revision这个时候可以稍微庆祝一下了，问题不大了，因为有修改就有可能。
具体怎么改就不多说了，10. Revision Submitted to Journal又开始了一个循环。
11. Accepted如果不要再审，只是小修改，编辑看后会马上显示这个状态，但如果要再审也会有上面的部分国外期刊投稿、审稿过程以及常用术语1. Author 作者如何在线投稿？在线投稿大致步骤：Step 1: Log In 登陆The login page gives you three options:1. Log in with your known User ID and Password 用户名和密码2. Check to see if you have an existing account 确认是否已经注册过3. Create a new account 没有就注册一个Step 2: Enter your Author Center 进入作者中心To begin a new submission, check a previous submission, continue a submission begun ear 认是新投，还是投修改稿Step 3: Inside Your Author Center 在个人的作者中心里面Existing manuscripts are found in one of three areas: 包括三个区域（这个每个杂志可能有Manuscripts to be Revised 需修改稿Partially Submitted Manuscripts 部分上传稿Submitted Manuscripts 已上传稿To start a NEW manuscript submission, choose “Submit First Draft of New Manuscript”Step 4: Entering Data 输入资料The following screens ask you to enter each piece of data associated with your manuscri manuscript, but needs to be entered in this format in order to make the system search notifications only. You cannot enter text into the Manuscript Data Summary table – s 提示一步一步输入Press “Save and Continue” at the bottom of each screen in order to save all of your wo your work will not be saved. 继续时选择保存和继续，如果点击back或者forward，原来输入Step 5: Upload Your Manuscript 上传文稿The File Manager is the area where you upload your files. Click on the Save and Conti Click on the Browse button in step #1. Locate your file and click on the name of the file t with the file name in step 浏览选定你电脑中要上传的稿件 #2. Select file is for review or must be made “yes” available for review in step #3. Click the blue upload button in Please refer to the “Author Instructions” for each specific journal to determine the Jo prior to submitting. You will not be able to change it once it’s submitted. 一定要看Yo ur original file will be stored and will be located under “Original Files / Files not will create files and place them under “Files for review.” You can make changes in s to your uploaded files. Click on "Save and Continue". 如果没有选择审稿，将不会被编辑看Step 5: Submit Your Manuscript 上传Click on “View uploaded files” - always view your proof carefully prior to submittin close the file, close the View uploaded files window and click on “Submit your manusc screen. You will also receive an e-mail confirmation that you can save for future refe 到你的邮箱。
E-Services: A Synthesis and Research AgendaCharles F. Hofacker1, Ronald E. Goldsmith1, Eileen Bridges2 and Esther Swilley1 This manuscript is currently in press (2007) in the Journal of Value Chain Management. This version is not paginated, and should not be used for citation purposes. A paginated version is available from the Journal of Value Chain Management.Notwithstanding any designation to the contrary in any printed version, copyright is held by the authors.Corresponding Author:Charles HofackerCollege of BusinessFlorida State UniversityTallahassee, FL 32306-1110 USA+1 850 644 7864chofack@1Florida State University, Tallahassee, FL, USA2Kent State University, OH, USAE-Services: A Synthesis and Research AgendaABSTRACTServices marketing research increases in both intensity and relevance as services contribute an increasing share of the world’s economy and as firms and their customers increasingly interact through electronic networks. E-services present sharp new challenges to both researchers and practitioners, because the processes from beginning to end of the e-service value chain are markedly different than those for offline services and because the electronic environment offers increased flexibility throughout the value chain. This flexibility creates the requirement to impose some sort of structure on all of the possible service and channel design choices. E-service flexibility creates an opportunity, and the need, to think about the consumer early in the design process. Finally, flexibility makes it difficult, but critically important, to consider various scenarios suggesting future developments in e-services. Our goal is to provide an overview of the past and some projections for the future in the new field of e-services.KEYWORDSE-Services, Flexibility, Self-Service Technologies, Service ProcessesINTRODUCTIONThe spread of electronic networks continues to transform business, marketing, and consumer behavior. One feature of this transformation is the appearance of the e-services phenomenon that arises at the border of two business domains of study: services marketing and e-commerce. Although some think about e-commerce in terms of the marketing of tangible goods (such as books, clothing, and electronics) online, a growing proportion of online activity is strictly devoted to the consumption of experiences; both hedonic and utilitarian.Of course, in the eyes of many observers, selling tangible goods online is itself an e-service, one which substitutes for physical retailing. It is therefore natural that services scholars would turn their attention to what is arguably the most important innovation in retail service delivery in many years, namely the ability to deliver service to a mass market with little to no direct human intervention. But the online world has seen a n ew category of service emerge which is neither a substitute for, nor a complement to, any offline offering. The emergence of purely virtual value chains, for example, in the form of Web portals, is frequently ignored in both e-commerce and services marketing texts (but see Fisk, Grove, and John 2003 for a notable exception). This new product category presents researchers with the need to develop and extend marketing theory, to investigate the new topics the theories suggest, and to formulate actionable recommendations for e-service managers. In order to further clarify the opportunities posed by e-services, we begin with some examples and a simple classification scheme.Three Prototypical Examples of E-ServiceWe discuss examples illustrating three extant types of e-services: (1) complements to existing offline services and goods, (2) substitutes for existing offline services, and (3) uniquely new core services.Organizations may add value to existing goods and services with complementary e-services. For instance, FedEx enhances the perception of tangibility while reducing both perceived and actual risk by letting customers track packages online. Major airlines allow passengers to conveniently change seats online. Cisco offers technical support and other after-purchase services on its website. As an additional benefit, use of this direct channel allows firms to unobtrusively collect information about their customers, thus permitting service improvements based on customer knowledge (Iqbal, Verma, and Baran 2003).Many firms utilize e-services as virtual substitutes for classic offline services. Retailers do this when they provide additional benefits online (e.g. more sizes or colors, longer product lines, backordering capability, and cost reduction). For instance, Amazon provides extensive availability compared to bricks-and-mortar bookstores, and offers value-added features, such as book reviews to assist in selection. Netflix makes it possi ble to search tens of thousands of movie titles and receive DVDs by mail along with pre-labeled return packages. Recommendation (Ansari, Essegaier, and Kohli 2000) and customization (Rust and Kanna n 2003) strategies are typically implemented more efficiently online than in person; they can generate significant competitive advantage both through lower cost and increased customer benefits.Finally, firms are developing new core e-services in the form of offerings that do not and generally could not exist as offline services. For instance, the online game provider World of WarCraft simultaneously hosts hundreds of thousands of gamers interacting in a shared virtual universe. Google Map's geographic service is available to an out-of-towner trying to find an address, to a business creating a real time parking spot exchange in Manhattan, to a government agency performing epidemiological analysis, and to a real estate broker offering an infomercial on l ocal housing offerings. Google Maps can provide a unique pattern of benefits with greater flexibility and more information than a traditional paper map. However, the benefits are limited by service design considerations as we note that directions provided in person may offer greater customization as well as responses to specific questions that may not be answered online.Complementing an offline service, substituting for an offline service, and offering a purely virtual core service are all identifiable as e-services. In all three cases service is provided by a programmed algorithm using network software rather than offered using human interaction. We now discuss related observations, beginning with a more formal definition of e-services.What Is an E-Service?E-services have previously been defined as "those services that can be delivered electronically," (Javalgi, Martin, and Todd 2004, p. 561) and similarly as "provision of services over electronic networks" (Rust and Kannan 2003, p. 38). Boyer, Hallowell and Roth (2002, p. 175) use the definition, "interactive services that are delivered on the Internet using advanced telecommunications, information, and multimedia technologies." The first two of thesedefinitions focus on the fact that delivery is electronic, and beg the question “what is a service?” or “what benefits are expected by the customer?” The third definition is concerned with the infrastructure necessary to deliver an e-service, but still does not define the term. Thus, it is important to clarify what we mean by “e-service” before we continue.Lovelock and Wirtz (2004, p. 9) define service as “an act or performance offered by one party to another…an economic activity that creates value and provides benefits for customers…by bringing about a desired change in, or on behalf of, the recipient.” This definition brings out both the process by which the service is produced and the outcome, in the form of benefits, that the customer receives. Both the service production processes and the outcomes are relevant when we consider e-services, as well.Regarding the service production process, an e-service is created and stored as an electronic code comprised of binary numbers, because it exists in a digital environment. Building on this, we observe that, by definition, the result of translating an act or performance into binary numbers is called an algorithm. Hahn and Kauffman (2002) have also identified e-services with algorithms. Using this idea, we could define e-service as: “an act or performance that creates value and provides benefits for customers through a process that is stored as an algorithm and typically implemented by networked software.” Thus our definition highlights the distinction between service production (a stored algorithm delivered by software) and service out come (the desired benefit received by consumers).We believe that the flexibility of algorithms and networked software combined with the requirement imposing structure on the service experience are distinguishing features of e-service which h elp to define the opportunities available to marketers. As an illustration, consider Yahoo!, which offers a calendar service to subscribers. The service production process begins with Yahoo! programmers who create and store the algorithms (procedures stored on computers which can be used to accomplish a task) that produce the calendar service. These algorithms can be programmed to behave in millions of different ways, producing different features, appearances, interactions and benefits, all of which might differ considerably from a physical calendar. Visually maintaining the metaphor of a paper calendar allows the consumer to bring her "calendar schema" to bear during the service experience. The benefit sought by a user of this service might be a reminder of an important birthday; thus, she creates a calendar entry for the date of the birthday either through her cell phone or Web browser. Before e-services were available, such a benefit might have been provided by a human personal assistant. Today the consumer might decide to have that Yahoo! reminder delivered to an email inbox or to a phone number, or she might receive the reminder through a cell phone, PDA, laptop or desktop machine. An e-service is logically independent of the devices that create, store, and deliver it. This logical independence of the service delivery process - perhaps we can dare to call it "separability" - creates an additional level of flexibility.Given the above process- and outcome-oriented definition of e-services, the question arises: what should services researchers study in the e-services value chain? We believe researchers should consider the design of the service process, any customer interaction required by the process as the service is consumed, and the outcome of the process, the latter directly reflecting the servicebenefits that the customer expects to receive. Emphasizing processes highlights flexibility and the arbitrariness of imposed structure in the design stage: e-service processes, as algorithms, may or may not be designed to imitate traditional service processes. In addition, they can be inventoried, repurposed, reassembled with different properties, recombined, customized, repackaged, re-branded, transferred or forwarded, delivered to various devices, and re-consumed. An e-service outcome, on the other hand, may be measured in the traditional way relative to customer expectations: satisfaction depends on reactions to any experiential aspects of the service delivery process and on the perceived results, as compared to expectations.We wish to continue comparing production processes and outcomes of e-services to those of more traditional services; thus we next consider properties of services and how they might apply to e-services. We then offer a selective review of the e-services literature, primarily to aid in identifying important themes. Based on this review, we highlight additional topics that in our view could benefit from a service-centric approach, but which have received only minimal attention from services researchers.PROPERTIES OF E-SERVICESWe proceed by discussing four commonly-cited properties of services and their managerial consequences: intangibility, heterogeneity, inseparability of production and consumption, and perishability (see Zeithaml, Parasuraman, and Berry 1985). In doing so we note that these traditional differences between services and goods have recently come under fire (see Lovelockand Gummesson 2004) because they do not offer a clear distinction between services an d goods, and because they represent a negative definition of services in terms of "what tangible goods are not" (Vargo and Lusch 2004b). For our purposes, however, they are useful in distinguishing e-services from more traditional services. Lovelock and Gummesson suggest identifying services as those purchases that do not result in ownership for the buyer; rather, benefits are obtained through access or temporary possession. This “nonownership” dimension is helpful in identifying e-services as a unique product category and we will use it also.IntangibilityWe might assume an e-service is less tangible than the same service delivered in person. In prototypical or pure cases of e-services, previous work on intangibility (i. e. Laroche, McDougall, Bergeron, and Yang 2004) will be especially relevant. In other cases, such as continuously delivered services (insurance), services which involve processing possessions rather than people (package delivery), or where there is an important symbolic component (plane reservation, ticket to a play) an e-service conveys increased tangibility.The specific delivery mechanism (DVD, kiosk, hand held device, personal computer) and format (web page, email, video, text message, voice menu) also offer an important contribution to tangibility. Although an e-service designer has considerable choice and flexibility in terms of delivery options, an e-service consumer has only a fixed set of senses and limited information processing capability. Thus, choice of channel should take into considerati on the specific type of intangibility (Laroche, Bergeron, and Goutaland 2001) and how that plays to the weaknesses and strengths of the human information processor.To elaborate further on possible weaknesses, if the consumer must deal with more than a few items of information simultaneously, a large computer screen might be called for instead of a cell phone screen or an auditory e-service. Otherwise the limitations of short-term memory (Bettman 1979) might reduce the benefits of the service. Similarly, the richness of the interaction afforded by the combination of the available bandwidth and the input and output devices determines the sorts of problem-solving possible in the mediated environment (Yadav and Varadarajan 2005), the amount and type of product complexity that can be conveyed by the seller and specified by the buyer, as it also dictates how compelling the brand experience is for the e-service. Interaction design likewise might need to take into account whether the benefit is reduction in uncertainty (adding data points) or reduction in ambiguity (Daft and Lengel 1986). Many of the strengths of consumer information processing grow over time with experience. Visual metaphors (Carroll and Thomas 1982) that honor schemas can leverage previous consumer knowledge while rendering the service process more tangible.HeterogeneityHeterogeneity represents variability in the quality and essence of a particular service. Given the error checking capabilities of networked software, an e-service is likely to be far more homogeneous than other services because it is not labor intensive (from the point of view of the firm), and so therefore does not incur as much risk of human error. In fact, upon production, an e-service is more homogeneous than a typical physical good. Rather than study heterogeneity in this sense, e-services researchers have focused on the impact of variability due to consumer participation (Parasuraman 2000), on customization and personalization (Goldsmith 1999), andon consumer heterogeneity in preferences for experience goods (Villas-Boas 2006). Variability is also added by the consumer's software configurations and preferences and hardware environment.Inseparability of Production and ConsumptionBecause "place" is not a property attributable to networked software (Kobrin 2001), e-services are highly flexible in terms of physical separation between consumer and producer. A musical band can record a song, which is an experiential product, and sell it on a website. We might determine that the service production (performing the song) and the service consumption (listening to it) have been separated in both space and time. Of course, the same song can be copied by the consumer to different media or played on an MPEG player or the car or home stereo system, or sampled and used in creating a new work of art. This example shows that the flexibility of an e-service can render it more separable than a physical good. However, if we consider an online music retailer that offers a v ariety of access methods, including downloading, we might classify that as a case where the consumer must be "present" on the website in order to consume the service.In either case, we believe that management's strategic goal should be to develop the ability to offer as many benefits as possible (Vargo and Lusch 2004a) by encouraging the sorts of transformations described above (e.g. copying, sampling), and by using updates, outtakes, newsletters, interactive chats with musical groups, and so forth.PerishabilityVargo and Lusch (2004b) argue that, in some cases, services are not perishable and can be inventoried. An e-service, being an algorithm, offers an excellent example of just such an exception, as it can be stored indefinitely by the firm (server disk) or consumer (CD or other media). We conclude that e-services are not necessarily perishable, as a consumer who has enjoyed a downloaded copy of The Iliad and the Odyssey might confirm. Unlike goods or offline services, binary numbers delivered by software can be consumed over and over again without being used up. Further, unlike offline services, an e-service such as the downloaded song mentioned above can be copied and given to someone else and yet still be retained. While offline services cannot be inventoried, e-services, as illustrated by song exchange, frequently have the opposite problem in that they are too easily inventoried, i.e., they are non-excludable in supply (Krishnan, Smith, and Telang 2003), meaning that management cannot prevent consumers from copying, storing and exchanging. Once again, a service-centric viewpoint helps to manage this problem. For example, game provider World of WarCraft gives away the PC software used to play the game, and charges players instead for access to the interactive server.NonownershipLovelock and Gummesson (2004, p. 34) propose that nonownership uniquely identifies services, which is to say that there is no transfer of ownership in services. This is true for both offline services and online services, although e-services have the additional characteristic of being non-rival in demand (Asvanund, Clay, Krishnan, and Smith 2004) meaning that consumption can occur simultaneously without reducing the other consumer's utility.Thus, e-services exist at the nexus of the intangible product and the use of software to perform functions previously carried out by humans. They embody the need satisfactions of traditional services, but using a unique technology. Before reviewing the current literature on e-services, we summarize the difference between goods, traditional services and e-services in Figure 1.--------------------------------Insert Figure 1 about Here--------------------------------E-SERVICE THEMESWe reviewed the e-services literature, considering theoretical, research and managerial issues raised for e-services by some of the traditional concerns of services marketing. The themes we discuss were selected because they addressed e-services in a marketing, e-commerce, or services context. Table 1 presents the framework for our discussion showing categories and sub-categories, some key articles, and a concise description of some key issues addressed.--------------------------------Insert Table 1 About Here--------------------------------We identify two frequent e-commerce application contexts, which marketers originally developed for offline services. First, much of e-services research focuses on online retailing, a substitute for a physical service. This is most clearly evident in work on e-service quality, historically an important research stream in offline services and now translated for the Web (e. g. Parasuraman, Zeithaml, and Malhotra 2005; Zeithaml, Parasuraman, and Malhotra 2003). A second and related point is that e-services are also frequently conceptualized as supplementary or peripheral services provided by companies whose main business is offline or whose valueproposition predates the Internet era. This is particularly evident in research on Self-Service Technologies (SSTs) (Dabholkar 1996) and Servicescapes (Bitner 1992). We now review these and other common themes that have emerged over the past few years. Our review of these e-service themes begins at the start of the e-service value chain with a discussion of e-service production; including multi-channel production, co-production, and self-service. In the second theme, the focus is maintained on back-office value-adding activities including service operations and fulfillment. In the third theme, we turn to more "customer facing" aspects of e-services, reviewing research on the servicescape and service quality. In the fourth theme we cover the classic services topics of e-service failure, recovery, satisfaction and loyalty. Our final theme concerns service relationship management.Who produces the service?Firms now have a variety of service channels at their disposal, which can be used alone or in combination with other channels. We begin this section by discussing e-service in the context of multi-channel service provision. In a multi-channel context, who produces the service becomes, at least partially, a strategic decision for the firm, endowing additional flexibility and potential for competitive advantage.We note that there us a disparity, or at least difference in emphasis, that is apparent in the background literature on "who produces a service." Prahalad and Ramaswamy (2004) and Vargo and Lusch (2004a) make a general argument in favor of allowing customers to co-create offerings with self-service technology. Specific benefits to the consumer of adding an Internet channel include customization, interactivity, global access, real time access and multimediadisplays (Viswanathan 2005). There are also strategic benefits to the firm: channel flexibility (the information mix, product representation, and site and interface design), market lock-in (switching costs), and positive consumption externalities (many e-services become more valuable as more consumers use them), according to Viswanathan (2005).We discuss externalities in the Virtual Communities section below, but for now, we need to acknowledge that the rosy picture painted by the above list of firm benefits is not universally shared. Bendapudi and Leone (2003) offer a cautionary view, pointing out that self-serving bias will lead to customers claiming more responsibility for a self-service success and taking less blame for a self-service failure. Similarly, self-service technologies reduce the impact of social bonds and feelings of social obligation (Selnes and Hansen 2001). There are important implications for loyalty in these cases. Surely more research is needed to determine exactly when and where co-production is beneficial, and how to mitigate its potential negative consequences for loyalty.Despite these cautionary notes, managers should reap a net benefit in the increased flexibility derived from having multiple service channels. Factors that trade off in choice of service channels include labor market availability or sh ortages (Rayport, Jaworski, and Kyung 2005), the amount of customer support desired and the cost of that support (Simons, Steinfield, and Bouwman 2002), the channel's physical capacity to perform various service outputs (Seiders, Berry, and Gresham 2000; Wallace, Giese, and Johnson 2004), the fit between the user's task and the channel software (Dishaw and Strong 1999), and the level of service product complexity (Boyer et al. 2002).Of course, consumers are not neutral with respect to how they deal with complexity; they bring their own preferences and abilities to this arena. Empirical research shows that the service channel features preferred by consumers vary greatly (Iqbal et al. 2003), adding to the difficulty of designing channel strategy. Consumers differ in self-efficacy (McKee, Simmers, and Licata 2006), need for social interaction (Dabholkar and Bagozzi 2002; Meuter, Ostrom, Roundtree, and Bitner 2000), readiness to accept self-service technology (Parasuraman 2000), and in their ability to perform the tasks demanded by it (e. g. Dellaert and Stremersch 2005). Both the cognitive limitations of consumers and the demands of branding tell us that consistency and seamlessness are important goals of management in a multi-channel service environment (Bendoly, Blocher, Bretthauer, Krishnan, and Venkataraman 2005; Sousa and Voss 2006).At this point we note an interesting disparity between the SST literature, which tends to focus on adoption (Bhappu and Schultze 2006; Lee and Allaway 2002; Meuter, Bitner, Ostrom, and Brown 2005), and the customer relationship management literature, which is retention-oriented. We believe it would be useful for scholars to contemplate the full SST life cycle, a topic we address later. For now, we continue to address the e-service value chain, turning to service operations and fulfillment.Service Operations and FulfillmentBecause e-service delivery is executed by software algorithms rather than human actors, the service development process differs from that of interpersonal services, resembling manufacturing operations more than classical service operations (Meyer and Zack 1996). Inaddition to being software-intensive, e-services are data-intensive, and data availability and accessibility, or data completeness (Brohman, Watson, Piccoli, and Parasuraman 2003), is a key driver of e-service value and convenience. It is the nature of information assets, including both algorithms and customer data, to be capable of being used, reused and recombined in various ways. For example, Google combines software and feedback from customer search input to create added value for Web surfers. It combines historical search input with data on click through probability to create added value for advertisers. Recombining these sorts of company assets is at the heart of what strategy theorists refer to as dynamic capabilities (Amit and Zott 2001). Companies offering new core e-services are among the most dynamic in the economy.For e-retailers, processes occurring behind the scenes can also be critical. We can divide online retail experience into pre- and post-sale time periods. Evidence is beginning to emerge of important recency effects; in particular, post-sale influences outweigh pre-sale factors in repurchase intention (Posselt and Gerstner 2005). The relationship between delivery waiting time, inventory policy, and operations decisions are examined in papers by Rabinovich (2004) and Cao and Zhao (2004). These provisions are important to the successful provision of e-services.In the business-to-business marketing sphere, e-service is about collaboration and relationship building. B2B e-service plays an important role in the trend towards supply chain integration and coordination (Bridges, Goldsmith, and Hofacker 2005; Iyer, Germain, and Frankwick 2004). B2B suppliers often create more powerful electronic fulfillment mechanisms than do B2C e-retailers, and elaborate Enterprise Resource Planning (ERP) systems (Gardner, Hanna, and。
Dear Dr. viaud,I read with interest the abstract of your article in Mutation Research, “Cyclophilin A and calcineurin functions investigated by gene inactivation, cyclosporin A inhibition and cDNA arrays approaches in the phytopathogenic fungus Botrytis cinerea”. Unfortunately, our library does not subscribe to Mutation Research. I would be most appreciative if you couldplease send me a .doc or .pdf file of the article by e-mail.Thank you,JiangPHD candidatePlant Disease Research Lab.,College of Plant Science and Technology,XX XX University,Wuhan, Hubei, 430070,P. R. China.Dear Dr. viaud,I'd appreciate a file (pdf) containing your article: Cyclophilin A and calcineurin functions investigated by gene inactivation, cyclosporin A inhibition and cDNA arrays approaches in the phytopathogenic fungus Botrytis cinerea, which appeared in Molecular Microbiology.(2003 Dec; 50(5): 1451-65.). Directly fishing out my thesis of the relative gene of the certain phytopathogenic fungus.Thank you in advanceBest wishesJiangPHD candidatePlant Disease Research Lab.,College of Plant Science and Technology,XX XX University,Wuhan, Hubei, 430070,P. R. China.Dear Dr. viaud,I am a post-doctoral researcher in Liu's lab at XX XX University.I have recently read your paper “Cyclophilin A and calcineurin functions investigated by gene inactivation, cyclosporin A inhibition and cDNA arrays approaches in the phytopathogenic fungus Botrytis cinerea”, which appeared in Molecular Microbiology.(2003 Dec; 50(5): 1451-65.). I am interested in trying your approach for detecting the same gene inanother fungus.I was wondering if you would be kind enough to send me the pETTthMutS plasmid reported in your paper.My mailing address is:XX XX@XX XX.Plant Disease Research Lab.,College of Plant Science and Technology,XX XX University,Wuhan, Hubei, 430070,P. R. China.Kind regards,JiangDear Dr. Andreas Schroeder, David Pauleen, Sid HuffI read with interest the abstract of your article in Inderscience Publishers, “Emerging evidence on linkages between Knowledge Management (KM) governance and management strategy: the case of two organisations”. Unfortunately, our library does not subscribe to Inderscience Publishers. I would be most appreciative if you could. please send me a .doc or .pdf file of the article by e-mail: ***************.Thank you in advanceBest wishesYaoweiBusiness school of nankai university94 Weijin Road, Nankai District, Tianjin 30071, P.R.China在课题研究中，难免碰到比较好的参考文章，但是学校图书馆却没有购买，此时向原作者索要原文是最好的办法。
投稿过程中的英文理解问题汇总decision pending是具体负责稿件的编辑已做出决定，发给主编确认阶段。
即从责任编辑转到主编手里，由主编决定。
说明：（1）将内容大体分几类，请大家按类查找。
（2）基本贴出了一些原文。
使用者可以使用搜索功能寻找自己的INTERSTING。
方法为使用网页工具栏中的：编辑——查找当前页。
（3）希望大家以后还有类似问题，在该贴后跟贴。
（4）不全之处，请其他战友补充。
1、1）Supply illustrations at the size they are to be printed, usually 76 mm wide (single column of text) or for especially large figures 161 mm (two columns of text). The intermediate width of 100 mm is also available should neither of these suffice.2）Do not put a box around graphs, diagrams or other artwork.3）Ensure that lettering is appropriately sized – should correspond to 8 or 9 pt when... amp;tpg=2&age=02、Conflicts of interestWe ask authors to state all possible conflicts of interest, including financial and other relationships. If you are sure that there is no conflict of interest, please state this. You might like to look at an editorial in the British Medical Journal on Beyond conflict of interest (cgi/content/short/317/7154/291). Remember that sources of funding should be acknowledged in your paper.... amp;tpg=2&age=03、Submission items include a cover letter (Authors are highly encouraged to include a list of 5-6 potential reviewers for their manuscript, with complete contact information), the manuscript (including title page, abstract, manuscript text, references, and table/figure legends), tables, and figures.The manuscript must be accompanied by a covering letter detailing what you are submitting (type of contribution, title, authors' names and affiliation, etc.).... amp;tpg=3&age=04、1）.Authors can submit their articles electronically via the “Author Gateway” page of this journal ( system automatically converts source files to a single Adobe Acrobat PDF version of the article, which is used in the peer-review process. Please note that even though manuscript source files are converted to PDF at submission for the review process, these source files are needed for further processing after acceptance.2）.Alternatively authors can submit by sending three hard copies of the manuscript and matching disk or e-mail directly to an editor who is an expert in the field of work being submitted.3）.The Publisher welcomes the receipt of an electronic version of your accepted manuscript (preferably encoded in LATEX). If there is not already a copy of this (on diskette) with the journal Editor at the time the manuscript is being refereed, you will be asked to send a file with the text of the accepted manuscript directly to the Publisher by e-mail or on diskette to the address given the electronic file is suitable for processing by the Publisher, the article will be published without rekeying the full text. The article should be encoded in LATEX, preferably using the Elsevier document class ‘elsart’, or alternatively the standar d document class ‘article’ or the document style ‘re-vtex’. ... amp;tpg=3&age=05、1）Manuscripts should be written in clear, concise form and sent to the Director，Three copies of the manuscripts must be written in English and double-spaced throughout.2） Each table should be titled, appropriately numbered and typed on a separate sheet.Units of measurement should be indicated and all abbreviations defined. 3） All the illustrations (graphs, drawings and photographs) should be referred to in the text as Figures. These should be good quality glossy photographs (or original India-ink drawings).4） Since graphs are generally reduced in size to cm, numbers, letters and symbols are to be originally large enough so that, when reduced, they will remain at least 2 mm high to maintain legibility. To avoid reduction, photographs' width should generally not exceed cm. When necessary, photographs can be arranged to form a plate of the maximum size of 17 cm (width) x 18 cm (height). Legends to figures should be typed on a separate sheet.5） Reprints should be ordered when returning the revised proofs and will be charged to the Authors.... amp;tpg=3&age=06、The Title page should include (1) a short and informative full articletitle (series titles are not accepted); (2) names of all authors (with one forename in full for each author), followed by their affiliations (department, institution, city with postcode, country); (3) the mailing address, fax and phone number and e-mail address of the corresponding author; (4) a running title of 50 characters or less.... amp;tpg=4&age=07、Entering a Comment is Required for Submission.You must include below an ethical statement that the manuscript has not and will not be submitted for publication elsewhere, while it is in review for MRB.Please suggest 3 suitable reviewers for your manuscript. Please provide the full name and e-mail address, and if possible, telephone number and postal address of each suggested reviewer.Please enter any additional comments that you would like to send to the editorial office. These comments do not appear in your manuscript. ... amp;tpg=4&age=08、running title，teaching cases... amp;tpg=5&age=09、Two complete sets of ullitrations must be submitted with legends typed on the same sheet... amp;tpg=6&age=010、All illustrations should accompany the typescript, but not be inserted in the textphotographs....should have the author's name, bthe figure number and an indication as to which is the top of the picture.Script should be typed doulbe-spaced on one side of the paper only", "doulbe-spaced“... amp;tpg=6&age=011、“The name of the author and the number of the figure (in Arabic numbers) should be written in the margin in blue pencil.”The legends of the figures should be typed on a separate sheet headed "Figure legends".）（figure legend）... amp;tpg=7&age=012、Submit the text (including any Tables) as a single file, in a standard disk inch, not high density; 100 and 250 MB Zip disk and CD are also accepted), together with two hard copies of the manuscript and two sets of figures (not photocopies). Not photocopies... amp;tpg=7&age=013、reviewer suggestions... amp;tpg=8&age=014、The manuscript must be accompanied by copies of all relevant papers published elsewhere by the authors and by copies of relevant manuscripts that are in press or under editorial consideration. If any tables or illustrations have been published elsewhere the editorial office must be informed so that permission to reproduce can be obtained from the original publishers.Papers must be arranged in the following order of presentation: title of paper; names of the authors; address of the place at which the work was carried out; a self-contained synopsis of the paper (100-200 words in length); tables; an abbreviated title for use as a running headline; captions to figures (on a separate page). Submissions not conforming to these guidelines may be returned to the authors.a self-contained synopsis of the paper (100-200 words in length); tables; an abbreviated title for use as a running headline; captions to figures (on a separate page)Figure captions should be submitted on a separate sheet of manuscript paper, clearly labeled. A total of 25 reprints of each paper will be provided free of charge to the author... amp;tpg=1&age=015、cover letter和reviewer suggestions... mp;tpg=17&age=016、In my experience, most overseas English journals with IF 1-5 don't require any review fees and publication fees. Moreover, you may receive up to 50 reprints for free when the paper is published.However, some (not all) journals with higher IF, you may need such fees. You have to check if before or when you submit (normally online) your paper.... mp;tpg=19&age=017、The authors should submit all the following files:official letter (paper with the heading of the institute, laboratory or company submitting the manuscript) signed by one of the authors, scanned as an image or pdf file.... mp;tpg=21&age=018、cover letter 和comment... mp;tpg=25&age=019、Running Title... mp;tpg=30&age=0【投稿状态】1、awaiting AE assignment... amp;tpg=3&age=02、with mini-hanging committee... amp;tpg=4&age=03、Current Status：Required Reviews Completed... amp;tpg=4&age=04、Awaiting Chief Editor Decision... amp;tpg=4&age=05、with referees Awaiting ED Recommendation... amp;tpg=5&age=0... amp;tpg=6&age=06、under review... amp;tpg=5&age=07、Required reviews completed... amp;tpg=5&age=08、with editor... amp;tpg=6&age=09、status date... amp;tpg=6&age=0awaiting reviwer assignment... amp;tpg=7&age=010、in press ；Corrected Proof；Available online 16 September 2005 ... amp;tpg=8&age=011、awaiting scores awaiting AE decision AE... mp;tpg=13&age=012、in review... mp;tpg=17&age=013、with editor for decision... mp;tpg=23&age=0【图表要求】1、Submit original figures with the final manuscript. Figures are normally glossy prints of photographs or crisp black and white reproductions (photostats) of line drawings. To submit figures as computer files follow the instructions below. Image resolution must be 300 ppi at final printed image size. If the final printed image size is unknown, size the image at a larger than final print size, maintaining at least 300 ppi resolution, and we will downsample the image to fit the final print dimensions (we cannot enlarge a digitized image). Image sizes as follows: 2175 pixels wide for a 2-column image; 975 pixels wide fora 1-column image.... amp;tpg=1&age=02、line artwork, halftone artwork, combination artwork(line/tone) ... amp;tpg=1&age=0... mp;tpg=13&age=03、“ Authors should NOT in addition then post a hard copy submission to the editorial office, unless you are supplying artwork, letters or files that cannot be submitted electronically, or have been instructed to do so by the editorial office.”“References to figures and tables should be made in order of appearance in the text and should be in Arabic numerals in parentheses, . (Fig. 2). Most file formats are accepted, but TIFF and EPS files, with fonts embedded, are preferred. If scanned, line art should be at a resolution of 800 dpi, and halftones and colour at 300 dpi. All colour values should be CMYK. ”... amp;tpg=2&age=04、Figures (photographs, diagrams and graphs)All illustrations should be provided in camera-ready form, suitable for reproduction (which may include reduction) without retouching. In addition, illustrations should be sent in electronic form wherever possible, with each illustration included as an individual TIFF or EPS file.Figures should be numbered consecutively in the order which they are referred to. They should not be included in the manuscript pages. The following standard symbols should be used as they are readily available to the typesetter: [] ע. Figure captions should be typed on a single sheet and placed at the end of the manuscript. The amount of lettering on a drawing should be reduced as far as possible by transferring it to the legend.... amp;tpg=3&age=05、Tables should be typed on separate pages and be numbered consecutively using Arabic numerals.... amp;tpg=4&age=06、For each item you wish to submit, scroll down and:Select the appropriate Item from the drop-down list. Mandatory Items are marked with an asterisk *.Enter a Description in the text box. Click Browse. In the open a window, select the file on your computer (original source file, not a PDF) and click Open. 'File Name' is filled now. Click Attach This File.Repeat steps 1-5 to attach the next submission Item When all Items have been attached, click Next at the bottom of the page.Note 1: Figures, graphics, photos should not be embedded in the manuscript text file. Please upload separate figure files, preferred formats are EPS, TIFF, JPEG.Note 2: If a Figure or Table is uploaded as a separate file, then please exclude this from the manuscript textfile, otherwise it will be shown in the PDF twice.... mp;tpg=15&age=07、Line graphs and bar charts should be sent in bitmap TIFF files witha minimum resolution of 1200 dpi, or as EPS files, with a preferred line width of 1 pt (minimum line width of pt). Bitmap images should be sent as TIFF files and not placed within EPS files .... mp;tpg=23&age=0。
英文投稿添加作者信件第一篇：英文投稿添加作者信件Dear Editor，Thanks very much for your kind work and consideration on publication of our manuscript entitled “论文标题”(ID ；论文编号).On behalf of my co-authors, we would like to express our great appreciation to editor and reviewers.Moreover, although this is a serious problem because the manuscript is already accepted for publication, we are wishing to add Dr.新增作者名字as an extra author to the manuscript and the new title page is sent as an attachment.新增作者名字 is a new member of my group and contributed much in the revised version ofour manuscript.After we received the referee's comments, he/she did a great jobfor the expansion of literature review so as to provide a deeper insight into the related published work and hence establish the contribution of this paper with respect to other publications tackling the problem.He/She also recalculated all the noise coefficients with a precision of four digits after the decimal point, as listed in TableII, TableIII, TableIV and TableVII.Without her hard work, we could not finish the revised version in several days.However, when I changed the authors' list,I forgot toadd her as the fifth author.Would you please to give usa chance to add Qin Gao as a new author to the paper? If you give us this chance, we will be grateful.Best wishes, Yours sincerely, XXXXX第二篇：英文信件格式英语信函写作大全超值篇书信是日常生活中常用的文体，是用以交涉事宜、传达信息、交流思想、联络感情、增进了解的重要工具。
Toward Spinozist Robotics:Exploring the Minimal Dynamics of Behavioural PreferenceHiroyuki Iizuka1,2and Ezequiel A.Di Paolo1 1Centre for Computational Neuroscience and Robotics,Department of Informatics,University of SussexBrighton,BN19QH,UK2Department of Media Architecture,Future University-Hakodate116-2Kamedanakano-cho,Hakodate,Hokkaido,041-8655,JapanContact author:Dr.Hiroyuki Iizukae-mail:ezca@sacral.c.u-tokyo.ac.jpphone:+44-1273-87-2948AbstractA preference is not located anywhere in the agent’s cognitive ar-chitecture,but it is rather a constraining of behaviour which is in turn shaped by behaviour.Based on this idea,a minimal model of behavioural preference is proposed.A simulated mobile agent is mod-elled with a plastic neurocontroller,which holds two separate high dimensional homeostatic boxes in the space of neural dynamics.An evolutionary algorithm is used for creating a link between the boxes and the performance of two different phototactic behaviours.After evolution,the agent’s performance exhibits some important aspects of behavioural preferences such as durability and transitions.This pa-per demonstrates1)the logical consistency of the multi-causal view by producing a case study of its viability and providing insights into its dynamical basis and2)how durability and transitions arise through the mutual constraining of internal and external dynamics in theflow of alternating high and low susceptibility to environmental variations.Implications for modelling autonomy are discussed.keywords:behavioural preference,homeostatic adaptation, dynamical systems approach to cognition,evolutionary robotics1IntroductionHow does an embodied agent develop a stable behavioural preference such as a habit of movement,a certain posture,or a predilection for spicy food? Is this development largely driven by a history of environmental contin-gencies or is it endogenously generated?Kurt Goldstein(1934)described preferred behaviour as the realization of a reduced subset of all the possible performances available to an organism(in motility,perception,posture,etc.) which are characterized by a feeling of comfort and correctness as a contrast to non-preferred behaviour which is often difficult and clumsy.Merleau-Ponty,following insights from Gestalt psychology,postulated that bodily habits are formed by resolving tensions along an intentional arc where the external situation solicits bodily responses and the meaning of theses solici-tations depend on the body’s history and dynamics.The overall tendency is thus towards an optimum or maximal grip on the situation(e.g.,likefinding just the right distance to appreciate a painting),(Merleau-Ponty,1962,p. 153).In these views,the fact that a preferred behaviour is observed more often would be derivative and not central to its definition(preferred be-haviour is often efficient but not necessarily optimal in any objective sense). Following this idea,we define a preference as the strength or commitment with which a behavioural choice is enacted,which is measurable in terms of its robustness to different kinds of perturbations(internal or external). Such a preference is typically sustained through time without necessarily being fully invariant,i.e.,in time it may develop or it may be transformed into a different preference.In order to understand such durable states from a dynamical systems perspective,it is therefore convenient to study under what conditions these preferences may change since this will reveal more clearly what are the factors that contribute to their generation.The word preference has many higher-level cognitive connotations that may not be captured by this minimal definition.On the one hand,we recog-nize that the dynamical systems picture portrayed in this paper does not do full justice to the richness of the concept(e.g.,human preferences involve a complex interaction between habits,needs,cultural context and sometimes conflicting values).On the other hand,our purpose is precisely to explore the minimal dynamical properties that might be shared by many instances of preferred behaviours.We follow the directions of synthetic minimalism which has been defended as a useful route towards clarifying complex ideas in cognitive science(Beer,1999;Harvey,Di Paolo,Wood,Quinn,&Tuci, 2005).Our objective is to achieve such a clarification of the term preference and related terms such as disposition,tendency,commitment,conation,etc.using the language of dynamical systems.Dynamical systems approaches to cognition have typically examined pro-cesses at the behavioural timescale such as discrimination,coordination,and learning,(e.g.,Beer,2003;Kelso,1995)and they have also been deployed to describe changes at developmental timescales(Thelen&Smith,1994). But of course,the strength of the approach lies in its potential to unify phe-nomena at a large range of timescales.In particular,behavioural preferences and their changes lie between the two scales just mentioned(the behavioural and developmental)and share properties with both of them.Goldstein has argued that we cannot reallyfind the originating factors of a preferred be-haviour purely in central or purely in peripheral processes,but that both the organism’s internal dynamics and its whole situation participate in de-termining preferences(Goldstein,1934).In this view,it becomes clear that a preference is never going to be captured if it is modelled as an internal variable(typically a module called“Motivation”)as in traditional and many modern approaches,but that a dynamical model needs to encapsulate the mutual constraining between higher levels of function,such as performance, and lower processes,such as neural dynamics(Varela&Thompson,2003).A preference is not“located”anywhere in the agent’s cognitive architec-ture,but it is rather a constraining of behaviour which is in turn shaped by behaviour.A goal of this work is to explore and possibly operationalize how this might be achieved in concrete terms suitable for further hypothesis generation.Preferences as described here are not typically addressed in minimally cognitive or robotic models.Most artificial agents are designed so as to have no preferences,or rather a single preference:that of adequate performance. Straying from the assigned task into a different behaviour chosen by the agent itself may be a sign of increased autonomy but not a frequent or explicit goal in current robotics.In this paper we present an exploratory model of behavioural preference with the objective of exploring the assertion of multi-causality.We consider that the minimal requirements to capture the phenomenon of preference is a situation with two mutually exclusive options of behavioural choice.An agent should be able to perform either of these options but the choice should not be random,but stable,and durable. The choice should not be invariant either but it should eventually switch(in order to study the factors that contribute to switching).There should be a correspondence between internal dynamical modes and stable behaviours. For this we use the methods of homeostatic adaptation to design not only the agent’s performance but to put additional requirements on the corresponding internal dynamics.We then examine the dynamical factors that play a rolein sustaining and changing preferences.2A Spinozist approachIn contrast to functional/computational approaches,a dynamical systems perspective on cognition makes it harder to conceptualize intentional terms (such as motivations,tendencies,goals,emotions,etc.)as functional states in the cognitive architecture of an agent(often implemented in computa-tional modules;a practice some people refer to as“boxology”).However, there isn’t as yet a clear and generally accepted alternative way to deal with intentional concepts from the dynamical camp.Attempts are not lack-ing.For instance,Kelso(1995)has suggested that it is possible to describe intentional behavioural changes in terms of transitions between dynamical attractors which correspond to different behaviours.The proposed process would be achieved through the successive stabilisation and destabilisation of attractors.In this view,intentions emerge as the order parameters of self-organised bodily and neural dynamics leading to behaviour when cou-pled with the environment.A similar idea has been proposed by Thelen and Smith(1994)who use the changes of attractor stability to explain the devel-opment of behaviours both at the developmental and behavioural timescales. In their view,the landscape of the stability changes depending on ontoge-netic processes.Juarrero(1999)offers a related view where prior intentions are understood as the setting up of a dynamical landscape of attractors(e.g., through recursive self-organizing activity on the system’s constraints)and proximate intentions imply the local selection of such dynamical alterna-tives.All these views share some problems,such as the problem of how the subject or agent to whom intentions might be attributed is itself dynamically constituted,i.e.,who is the agent that re-shapes a dynamical landscape so that we may speak of intentions as belonging to it.Nevertheless,the central intuition of dynamical re-shaping is appealing and worthy of expansion and clarification.An important point shared by both intentional behaviour for Kelso and Juarrero and developmental processes for Thelen and Smith is that they cannot be separated from the ongoing behavioural dynamics,i.e.,the sys-tems engagement with the environment.Given that these are processes that themselves may alter the dynamics of interaction with the environment,the emerging picture is one of mutual modulation,or circular causality(Clark, 1997;Thompson&Varela,2001)where intentional and developmental pro-cesses re-parameterize behaviour and,in turn,interaction with the environ-ment constrains and modulates intentions and development.Dynamical views similar to those of Kelso and Juarrero have been ex-plored in robotics.For instance,in the work of Ito et al.a humanoid robot shows transitions between two different ball handling behaviours using a special kind of neurocontroller in which specific nodes are trained to have an association with each behaviour(Ito,Noda,Hoshino,&Tani,2006). Transitions are demonstrated to occur when the robot is interrupted by a person changing the position of the ball.Although the functioning of the system may be conceived in terms of a re-organization of the attractor land-scape,transitions are demonstrated through human guidance,i.e.,triggered externally.This,unfortunately,does not allow the evaluation of more spon-taneous forms of switching in terms of the dynamical account of mutual constraining between interactive and neural levels that we want to explore in this paper.Following this idea,in this paper we suggest that a way of approaching a study of preference is by embedding the circular causality between internal organization and interaction in an embodied agent through the application of homeostatically-driven neural plasticity(further justified below).Even though this idea is not disconnected from those of Kelso,Juarrero and The-len and Smith,the view in our proposed approach is different from the attractor model in the following important sense.Suppose that a dynamical system has two attractors,each of which cor-responds to a different behaviour.If a transition between them occurs,it could only be caused when noise or perturbations to the dynamics are strong enough to detach the trajectory from the current attractor.As Kelso dis-cusses,this would not constitute a proper intentional change because the behavioural transition depends only on a random event and would not re-sult in durable behaviour as in the case of a preference.Conversely,the system cannot change attractors without noise,which means that in such a case it would produce a same behaviour permanently.Therefore,there are problems in explaining preferences in behaviour as the switching between attractors in the proper dynamical systems sense. It should be noted that we are not denying noise and random events may play a role in triggering the transitions but insisting on the significance of transitions caused by internal factors,which can be seen as higher-level dynamics as in Kelso’s and Thelen and Smith’s explanations.A proper intention,motivation,or alteration to a preference should also be dependent on endogenous conditions,such as changes to the dynamical landscape itself as a result of a history of behavioural interactions with the environment in non-arbitrary ways.By explicitly implementing the circular causalitybetween inner stability and external behaviour into our model through the use of homeostatic mechanisms(as only one possibility for achieving this), important aspects of preferences such as durability and transitions between behaviours can be captured.However,the choice of homeostatic adaptation to model the dynamics of preference still requires a stronger justification.A view on preferences,dispositions and tendencies that is amenable to a dynamical interpretation is Spinoza’s doctrine of conatus or striving de-veloped in his Ethics(IIIp4-7).This view serves as an inspiration for how these issues are addressed in our model.We read in part III proposition 6:“Each thing,as far as it can by its own power,strives to persevere in its being”.There are several issues concerning this doctrine discussed by Spinoza scholars,especially the unargued proposition IIIp4that states that “No thing can be destroyed except through an external cause”(Matson, 1977);one might think of a bomb as a counter-example.However,in the context of cognition a dynamical systems approach is highly compatible with views that seek to establish a continuity between life and mind,(Jonas,1966; Maturana&Varela,1980;Stewart,1992;Wheeler,1997;Bourgine&Stew-art,2004;Weber&Varela,2002;Di Paolo,2005).The essence or being of a living system is its self-producing organization(autopoiesis)and it so happens that IIIp4is indeed true of living systems at least at a minimal level of organization,(Di Paolo,2005)–even though this is not such an ob-vious statement for more complex systems beyond bare autopoiesis;think of autoimmune diseases.Hence,we may assume that the striving doctrine is applicable to minimal cognitive systems as well and so we can use this idea as a regulative principle informing our approach(complex cases such as culturally-embedded human cognition may present special kinds of prob-lem such as the origin of conflicting values;we are not considering these possibilities here).So the question of preference becomes the question of understanding an agent’s changing conatus.For Spinoza,this would directly relate to the agent’s current being.A hungry animal strives for food and a thirsty one for water.Conatus resolves in the interactional domain(actions and perceptions)a tension that originates in the internal constitutive domain (an internal need).To cash out such ideas in dynamical terms we must define in the dynamical organization of the agent a condition of tension and satisfaction where thefirst is defined as being in a state that leads to alterations in the system’s organization(its essence or being in Spinozist terms)and the second as being in a state where the system’s organization remains invariant.This closely follows Ashby’s idea of an ultrastable system(Ashby,1960).Conatus is therefore,on afirst approximation,the interactive and internal striving of an ultra-stable system to remain in those states that satisfy the condition of invariant organization.This need not be interpreted teleologically yet even though the terms tension and satisfaction have not been chosen innocently,(for a discussion regarding the conatus doctrine and teleology see(Bennett,1984,1992;Curley,1990)).This view will be refined after understanding the results from our model.Finally,in order to sharpen the contrast of the view we are proposing to previous ones,a relation to action selection should be mentioned.The action selection literature is concerned with how animals,robots or simu-lated agents can solve the ongoing problem of choosing what to do next in order to achieve their objectives(Humphrys,1997;Bryson,2003).Agents are given several options of action units in advance and they must decide which one should be taken next and how they should be ordered and com-bined.In other words,in action selection models the sensorimotor coupling at the lower level isfixed and discretised in order to set the action units and the aim is to decide how they should be chosen for the planning of the higher-level goals.Roughly,this means that planning as a higher-level de-scription is separable from the ongoing actions as a lower-level description although this may depend on how the action units are defined.It might be rather precise to say that this kind of work tries to illuminate our planning intelligence in an ecological multi-objective context at the cost of simplifying (even removing)the effects of the sensorimotor dynamics on the system’s changing organization.This can be seen even in work on embodied robotics which models intentional terms such as preferences,motivations and goals as certain variables or“compartmentalized”functions named emotions or motivations(Breazeal(Ferrell),1998;Vel´a squez,1997).Our view is differ-ent from the idea expressed in most action selection models in that planning is not separable from actions.Producing actions based on sensorimotor coupling organizes and preserves plans,and in turn plans regulate the sen-sorimotorflows.It should be clear that our motivations are also different although we would expect that future developments of our work may interact more closely with problems in action selection.3Homeostatic adaptation in neural controllersA salient feature of the homeostatic adaptation model proposed in(Di Paolo, 2000)is that local plastic adaptive mechanisms work only when neural acti-vations move out of a prescribed region;an idea inspired in Ashby’s home-ostat(Ashby,1960).Such a mechanism has been implemented in a neuro-controlled simulated vehicle evolved with afitness function rewarding pho-totaxis and the maintenance of neural activations within the homeostatic region.The use of intermittent plasticity in combination with this selec-tive pressure allows for the evolution of a novel kind of coupling between internal and environmental dynamics.Once the neurocontroller gives rise to behavioural coordination within a given environmental situation that re-sults in internal stability,synaptic weight changes no longer happen.If the situation changes,such as in an inversion of the visualfield or some other perturbation,this causes a breakdown of coordination,which means that the neural activations cannot in general be maintained within the homeo-static region.As this happens,the local adaptive mechanism is activated until itfinds a new structure(synaptic weight values)which can sustain the activations within the homeostatic region and(very likely,though not nec-essarily)re-form the behavioural coordination.As a result,the agent can adapt to perturbations it has never experienced before.This approach aims at creating a high dimensional bounded set,or box, in phase space that corresponds to neural homeostasis often linked to suit-able performance.The dynamics within the box is stable in that if trajecto-ries go out of the bounds,the network’s own configuration,by design,will change plastically until they come back again into the box.If trajectories remain within the box,the system’s configuration no longer changes.This plasticity-dynamics relation makes it possible for a coordination to occur (under suitable circumstances)between a higher function,phototactic be-haviour in this case,and the process that regulates the sensorimotorflows. If the behaviour cannot be achieved,homeostatic adaptation attempts to find new sensorimotorflows.This is therefore a concrete example of the circular constraining between levels mentioned above1.In the current context,we extend this property to create a model ofbehavioural preference.Our idea is that if the system holds two separate high dimensional boxes in the space of neural dynamics which are associ-ated with performing different behaviours,for simplicity’s sake phototaxis towards different light sources A and B,a preference could be formed by the dynamical transitions that select which box the dynamics go into and stay in.This provides one of our requirements for talking about prefer-ence,that of durability(bottom-up construction of the stability).Once a behaviour is formed,due to the stability in a box,the system keeps doing the behaviour while ignoring other behavioural possibilities.This plays the role of a spontaneous top-down constraint that regulates the sensorimotor flows.However,some disturbances might eventually cause a breakdown of the stability and then another behaviour can be reconstructed through the homeostatic adaptive mechanisms.Since by design,the system has an-other region of high stability,the system will be likely to switch into it and then start enacting its other behavioural option.In this way,behaviour can switch due to the corresponding transitions between two boxes.We expect to see both spontaneous and externally-induced transitions from the view-point of the top-down and bottom-up construction or destruction of durable but impermanent dynamical modes.Here wefind our second requirement, that of the possibility of transformation,or change in preference.Except for considerations of symmetry,the placement of these high-dimensional boxes is arbitrary in the present model(see discussion).4ModelOur proposed minimal model of behavioural preference extends the home-ostatic adaptation model.The idea is implemented in a simulated mo-bile agent with a plastic neural controller containing two separated,high-dimensional homeostatic regions.The simulated robot is faced with two different kinds of light as mutually exclusive options of behavioural choice. It must visit one of them and this“choice”must correspond to the internal dynamics being contained in the corresponding homeostatic box.An evolu-tionary algorithm is used to design the neurocontroller.Agents are evaluated on each type of light separately,and on both types simultaneously,in which case one of them is presented as blinking.The agent must approach the non-blinking light,(see below).After evolution,this cue is removed and the agent’s preference is investigated by presenting two constant,non-blinking lights of each type simultaneously.Agent.An agent is modelled as a simulated wheeled robot with a circular body of radius4and two diametrically opposed motors.The motors can drive the agent backwards and forwards in a2-D unlimited plane.Agents have a very small mass,so that the motor output directly determines the tangential velocity at the point of the body where the motor is located.The translational movement of the robot is calculated using the velocity of its center of mass,which is simply the sum of the two tangential velocities and the rotational movement is calculated by dividing the difference of the two motor outputs by the diameter(leading to a maximum translation speed of2.0units and rotational speed of0.25radians at each Euler time step). The agent has two pairs of sensors for two different light sources,A and B, mounted at angles ofπ/3radians to the forward direction.The two lights do not interfere with each other and the model includes the shadows produced by the agent’s body.Plastic controller.A fully connected continuous-time recurrent neural network(CTRNN)(Beer,1990)is used as the agent’s controller.The time evolution of the states of neurons is expressed by:τi˙y i=−y i+Nj=1w ji z j(y j)+I i,z i(x)=1/(1+e−x−b i),(1)where y i represents the cell potential of neuron i,z i is thefiring rate,τi (range[0.4,4])is its time constant,b i(range[−3,3])is a bias term,and w ji (range[−8,8])is the strength of the connection from the neuron,j,to i.I i represents the sensory input,which is given to only sensory neurons.The number of neurons,N,is set to10in this paper,4of which are assigned to sensory neurons,i.e.,2neurons for each light sensor.The sensory input is calculated by multiplying the local light intensity by a gain parameter (range[0.01,10]).The decay of the light intensity follows a sigmoid function in order to avoid the situation in which it becomes too strong when the agent is very close to the light.There is one effector neuron for controlling the activity of each motor.Similarly,the motor output is calculated from the firing rate of the effector neuron,which is mapped into a range[−1,1]and is then multiplied by a gain parameter(range[0.01,10]).All free parameters are determined genetically.A plastic mechanism allows for the lifetime modification of the connec-tion weights between neurons.The homeostatic regions are described by a plasticity function of thefiring rate of the post-synaptic neuron;this func-tion determines the strength of change of all incoming weights.The plastic function is0in the homeostatic regions,which means no plasticity.To as-sign the different phototactic behaviours to different boxes in our extended model,two separated regions are arbitrarily set corresponding tofiring rates of[0.15,0.4]and[0.6,0.85]as shown in Fig.1and for each neuron each region is arbitrarily assigned for phototaxis A or B at the start of the evolutionary run.To reduce bias,the upper(bottom)region of half of the internal neu-rons is assigned for phototaxis A(B),and the other region is for phototaxis B(A).These assignments remain the same for the evaluations during the evolutionary run.However,the scheme with two separate regions is not applied for input and output neurons because this would introduce biases in the input sensitivity as well as prescribe particular styles of movement.For input and output neurons,a function which has a single homeostatic region is applied(Fig.1(right)).Weight change follows a Hebbian rule which also depends linearly on thefiring rate of the pre-synaptic neuron and a learning rate parameter. Weights from neuron i to j are updated according to:∆w ji=ηji z i p(z j)(2) where z i and z j are thefiring rates of pre-and post-synaptic neurons,re-spectively,∆w ji is the change per unit of time to w ji,p(x)is the plastic function(see Fig.1),andηji is a rate of change(range[0,0.9]),which is genetically set for each connection.For simplicity,we restrict this parameter to a positive range,so that the product of this value and the plastic func-tion always works in the direction of returning theflow into the homeostatic region.For example,if thefiring rate of neuron j,z j,is between[0,0.15]or [0.5,0.6],the plastic function returns a positive value and the weight change is calculated by multiplying thefiring rate of pre-synaptic neuron i andηji, which are positive.This means that the weight change works to increase the firing rate z j.If thefiring rate is between[0.4,0.5]or[0.85,1.0],the plastic function returns a negative value.In this case,the weight change works in the opposite direction to decrease thefiring rate.This implementation for the plasticity rule is not the only possible way to realize neural homeostasis.Other studies show different forms of neural plasticity(Di Paolo,2000;Williams,2004).However,we use the simplest implementation for out current purposes in that plasticity always works in the direction of stability within the boxes.The validity of stipulating the homeostatic regions in this way will be addressed later in the discussion.The time evolution of agent’s navigation and plastic neural network are computed using an Euler method with a time step of0.1.。
authors and their affiliation全文共四篇示例，供读者参考第一篇示例：【作者和他们的机构】在学术领域中，作者和他们的机构通常被广泛讨论和关注。
通过作者的机构信息，可以更好地了解作者的研究背景、学术地位以及研究成果的可信度。
本文将探讨作者和他们的机构之间的关系，并讨论这种关系在学术界中的重要性。
作者的机构通常指的是作者所归属的机构、组织或研究单位。
这些机构可以是大学、科研院所、研究中心、实验室等地方。
作者的机构信息通常会出现在其发表的学术论文、书籍或文章的第一页或结尾处。
通过作者的机构信息，读者可以了解作者的工作背景和从事研究的领域，进而对作者的研究成果有一个初步的了解。
作者的机构可以影响其在学术界的声誉和地位。
一般来说，来自知名机构的作者往往会受到更多的关注和尊重。
这是因为知名机构通常拥有更好的学术资源和研究条件，能够吸引更多的顶尖研究人员，并进行更高水平的研究。
来自知名机构的作者往往会被视为更具有权威性和可靠性。
作者与其所属机构之间的关系也会影响到作者的学术发展和研究成果。
一般来说，作者与其机构的合作关系越密切，其研究成果也会越好。
作者与其所属机构的合作可以为作者提供更多的学术资源和支持，帮助其开展更高水平的研究。
作者的机构也会受益于作者的研究成果，提升机构的学术声誉和地位。
作者和他们的机构之间的关系也受到一些偏见和争议。
有时候，作者可能会因为所属机构的名声而受到错误的评价和判断。
有些人可能会认为来自知名机构的作者一定研究水平更高，这种看法是不正确的。
实际上，作者的学术成就主要取决于其个人的努力和能力，而不仅仅是所属机构的名声。
作者和他们的机构之间的关系是学术界中一个重要的研究领域。
通过研究作者和他们的机构之间的关系，可以更好地了解作者的学术地位和研究成果的可信度。
作者和他们的机构之间的紧密合作关系也能够促进学术发展，推动科学研究的进步。
作者和他们的机构之间的关系应该得到更多的关注和研究，为学术界的发展做出更大的贡献。
英文博客范文The Power of Blogging: Sharing Your Passion with the WorldIn today's digital age, the art of blogging has become a powerful tool for individuals to express themselves, share their knowledge, and connect with a global audience. Whether you're an aspiring writer, a passionate hobbyist, or a professional seeking to establish your online presence, the world of blogging offers a vast and dynamic platform for your voice to be heard. In this blog post, we will explore the transformative power of blogging and why it has become an essential component of modern communication.At its core, blogging is the act of creating and sharing content online. It allows individuals to express their thoughts, ideas, and experiences in a way that resonates with their readers. Unlike traditional media, where content is often controlled and filtered, blogging empowers individuals to take control of their narrative and share their unique perspectives with the world. This freedom of expression has given rise to a diverse tapestry of blogs, each offering a distinct and captivating perspective on a wide range of topics.One of the most compelling aspects of blogging is its ability to foster a sense of community. Through the interactive nature of blog platforms, readers can engage with the author, share their own thoughts and experiences, and even build meaningful connections with like-minded individuals. This interactive dynamic creates a rich and vibrant ecosystem where ideas are exchanged, debates are sparked, and relationships are forged. By tapping into this community-driven aspect of blogging, authors can not only share their content but also cultivate a loyal following of engaged and passionate readers.Moreover, blogging has become an invaluable tool for personal and professional growth. For individuals seeking to establish their expertise or build a personal brand, a well-crafted blog can serve as a powerful platform to showcase their knowledge, skills, and unique perspectives. By consistently producing high-quality content, bloggers can position themselves as thought leaders in their respective fields, attracting the attention of potential employers, clients, or collaborators. This can lead to new opportunities, increased visibility, and the ability to influence and inspire others within their industry.In addition to its impact on personal and professional growth, blogging has also become a powerful tool for businesses andorganizations. Companies have recognized the value of maintaining a strong online presence and using blogs as a means to connect with their target audience, share industry insights, and promote their products or services. By creating engaging and informative blog content, businesses can establish themselves as trusted authorities in their field, build brand loyalty, and drive traffic to their websites.Furthermore, the rise of blogging has also had a profound impact on the world of journalism and content creation. Traditional media outlets have had to adapt to the changing landscape, incorporating blogging and social media into their content strategies to remain relevant and competitive. Bloggers have emerged as influential voices, often breaking news, providing in-depth analysis, and offering alternative perspectives that challenge the status quo. This convergence of traditional and new media has led to a more diverse and dynamic information landscape, where readers can access a wide range of perspectives and stay informed on the issues that matter most to them.However, the power of blogging extends far beyond the realms of personal expression, professional development, and business growth. Blogging has also become a powerful tool for social change and advocacy. Bloggers have used their platforms to raise awareness about important issues, advocate for marginalized communities, and inspire collective action. By sharing personal stories, providingfactual information, and mobilizing their readers, bloggers have played a crucial role in driving social and political movements, fostering greater understanding, and promoting positive change in the world.In conclusion, the power of blogging lies in its ability to empower individuals, foster community, and drive meaningful change. Whether you're an aspiring writer, a passionate hobbyist, or a business seeking to connect with your audience, the world of blogging offers a vast and dynamic platform for your voice to be heard. By embracing the transformative potential of blogging, you can not only share your unique perspective with the world but also contribute to the rich tapestry of human expression and collective progress. So, what are you waiting for? Start your blogging journey today and let your voice be heard.。
英语作文关于博客网络日志The Rise of Blogging: A New Era of Online Expression。
In recent years, the rise of blogging hasrevolutionized the way people share their thoughts and experiences with the world. With the advent of the internet, individuals now have the ability to create their own personal online journals, known as blogs, where they can express themselves freely and connect with a global audience. This new form of online expression has given rise to a diverse and vibrant community of bloggers, who usetheir platforms to share their passions, insights, and personal stories.One of the most remarkable aspects of blogging is its accessibility. Unlike traditional forms of media, such as newspapers and magazines, anyone with an internetconnection can create and publish a blog. This has democratized the process of sharing information and ideas, allowing individuals from all walks of life to have a voicein the digital space. As a result, the blogosphere has become a rich tapestry of diverse perspectives and experiences, making it a valuable resource for those seeking new insights and perspectives.Another key feature of blogging is its interactive nature. Unlike traditional media, which is typically a one-way communication channel, blogs allow for real-time interaction between the author and their audience. Readers can leave comments, ask questions, and engage in conversations with the blogger, creating a sense of community and connection. This level of engagement has transformed the way people consume and interact with content, fostering a more dynamic and participatory online culture.Furthermore, blogging has provided a platform for individuals to build their personal brands and pursue their passions. Many bloggers have leveraged their online presence to establish themselves as experts in their respective fields, attracting a loyal following and even monetizing their content through advertising andsponsorships. This has opened up new opportunities for creative expression and entrepreneurship, empowering individuals to turn their hobbies and interests into viable careers.In addition to personal blogs, there are also a growing number of niche blogs that cater to specific interests and communities. From fashion and beauty to travel and food, there is a blog for virtually every topic imaginable. This has created a space for like-minded individuals to connect and engage with content that resonates with their interests, fostering a sense of belonging and camaraderie.However, despite the many benefits of blogging, there are also challenges and concerns that come with this new form of online expression. The rise of fake news and misinformation has raised questions about the credibility and reliability of online content, leading to calls for greater transparency and accountability among bloggers and content creators. Additionally, the pressure to maintain a consistent and engaging online presence can take a toll on the mental health and well-being of bloggers, leading toburnout and exhaustion.In conclusion, the rise of blogging has ushered in a new era of online expression, empowering individuals to share their thoughts, experiences, and passions with a global audience. The accessibility and interactivity of blogs have transformed the way people consume and engage with content, creating a vibrant and diverse online community. While there are challenges and concerns that come with this new form of media, the impact of blogging on the way we communicate and connect with one another is undeniable. As we continue to navigate this digital landscape, it is important to uphold the values of transparency, authenticity, and responsible content creation, ensuring that blogging remains a force for positive change and meaningful expression in the digital age.。
Bibtex使⽤介绍BibTeX 是⼀种格式和⼀个程序, ⽤于协调LaTeX的参考⽂献处理.BibTeX 使⽤数据库的的⽅式来管理参考⽂献. BibTeX ⽂件的后缀名为 .bib . 先来看⼀个例⼦@article{name1,author = {作者, 多个作者⽤ and 连接},title = {标题},journal = {期刊名},volume = {卷20},number = {页码},year = {年份},abstract = {摘要, 这个主要是引⽤的时候⾃⼰参考的, 这⼀⾏不是必须的}}@book{name2,author ="作者",year="年份2008",title="书名",publisher ="出版社名称"}说明:1. 第⼀⾏@article 告诉 BibTeX 这是⼀个⽂章类型的参考⽂献. 还有其它格式, 例如 article, book, booklet, conference, inbook,incollection, inproceedings, manual, misc, mastersthesis, phdthesis, proceedings, techreport, unpublished 等等.2. 接下来的"name1", 就是你在正⽂中应⽤这个条⽬的名称.3. 其它就是参考⽂献⾥⾯的具体内容啦.⼤概都是由这些⼀个个的这样的标签组成，每个表⽰⼀个⽂献，全部按照这样写即可。
@ARTICLE @CONFERENCE 这些指定了⽂献的类型，article是期刊⽂章，conference是会议⽂章，可以从名称上分辨。
类型有下⾯⼏种（来⾃ /documentation/LaTeX/reference/faq/bibstyles.html）：@article An article from a journal or magazine@book A book with an explicit publisher@booklet A work that is printed and bound, but without a named publisher or sponsoring institution@conference The same as inproceedings@inbook A part of a book, which may be a chapter (or section or whatever) and/or a range of pages@incollection A part of a book having its own title@inproceedings An article in a conference proceedings@manual Technical documentation@mastersthesis A Master's thesis@misc Use this type when nothing else fits@phdthesis A PhD thesis@proceedings The proceedings of a conference@techreport A report published by a school or other institution, usually numbered within a series@unpublished A document having an author and title, but not formally published@collection Not a standard entry type. Use proceedings instead.@patent Not a standard entry type.不翻译了，很简单的⽂字。
每天查询反链，搜索footprint、profile资源的命令大家肯定用的得心应手，但是大家肯定不知道有多少谷歌搜索命令，下面转来一起学习很多不常用的，但是说不定哪天用到了标记一下一、allintitle:当我们用allintitle提交查询的时候，Google会限制搜索结果仅是那些在网页标题里边包含了我们所有查询关键词的网页。
例［allintitle: detect plagiarism］，提交这个查询，Google仅会返回在网页标题里边包含了‖detect‖和‖plagiarism‖这两个关键词的网页。
二、bphonebook:用bphonebook进行查询的时候，返回结果将是那些商务电话资料。
三、allinanchor:anchor 是一处说明性的文字，它标注说明了这个链接可能跳转到其它的网页或跳转到当前网页的不同地方。
当我们用allinanchor提交查询的时候，Google会限制搜索结果必须是那些在anchor文字里包含了我们所有查询关键词的网页。
例[ allinanchor: best museums Sydney ] ,提交这个查询，Google仅仅会返回在网页anchor说明文字里边包含了关键词‖best‖―museums‖和‖Sydney‖的网面。
四、allinurl:当我们用allinurl提交查询的时候，Google会限制搜索结果仅是那些在URL（网址）里边包含了我们所有查询关键词的网页。
例［allinurl: google faq］，提交这个查询，Google仅会返回在URL里边包含了关键词‖google‖和‖faq‖的网页，象/help /faq.html等的网页。
五、cache:提交cache:url ，Google会显示当前网页的快照信息，从而替换网页的当前信息。
例［cache:］，提交这个查询，Google会返回所有抓取的关于的网页快照信息。
在显示的网页快照信息里边，Google会高亮显示查询关键词。
介绍几个网页的英文作文英文回答：Web pages are essential components of the internet, providing a way to share information, create online experiences, and connect with people around the globe. They are composed of a variety of elements, including text, images, videos, and interactive features, and are designedto be accessed and viewed through a web browser.The structure of a web page is typically defined by Hypertext Markup Language (HTML), which provides the framework for organizing content and creating the visual appearance of the page. HTML elements, such as headings, paragraphs, and lists, are used to define the structure and hierarchy of the content, while CSS (Cascading Style Sheets) is used to control the visual presentation of the page, including fonts, colors, and layout.Web pages can be static, meaning they remain the sameeach time they are accessed, or dynamic, meaning their content changes based on user input or external factors. Dynamic web pages are often created using server-side programming languages, such as PHP or Python, which allow for the creation of interactive and personalized experiences.The content of a web page can vary widely, depending on its purpose. Some common types of web pages include:Home pages: These pages provide an overview of a website, often including links to other sections of the site, such as about us, products, and services.Product pages: These pages showcase specific products or services, providing detailed information, images, and purchasing options.Blog posts: These pages typically feature written content on a particular topic or subject, often with the ability for readers to leave comments and interact with the author.News articles: These pages present news stories,current events, and other timely information.Contact pages: These pages provide contact information, such as addresses, phone numbers, and email addresses, fora business or organization.Web pages can also incorporate a variety of interactive elements, such as forms, surveys, and polls, which allow users to provide input and interact with the content. These elements can be used to collect feedback, gather data, and create more engaging experiences for users.In summary, web pages are versatile and dynamic components of the internet, providing a way to share information, create online experiences, and connect with people around the globe. They are designed using HTML, CSS, and other programming languages, and can vary widely in content and purpose.中文回答：什么是网页？网页是互联网中不可或缺的组成部分，它提供了一种共享信息、创建在线体验和与全球各地的人们联系的方式。
Author Guidelines for the Preparation of Contributionsto Springer Computer Science ProceedingsAlfred Hofmann 1,*{alfred.hofmann,ralf.gerstner,anna.kramer , Ralf Gerstner 1, Anna Kramer 1, and Frank Holzwarth 21Springer-Verlag, Computer Science Editorial, Heidelberg, Germany}@2 Springer-Verlag, Technical Support, Heidelberg, Germanyfrank.holzwarth@ Abstract. The abstract is a mandatory element that should summarize the con-tents of the paper and should contain at least 70 and at most 150 words. Ab-stract and keywords are freely available in SpringerLink.Keywords: We would like to encourage you to list your keywords here. They should be separated by commas.1 IntroductionYou will find here Springer’s guidelines for the preparation of proceedings papers to be published in one of the following series, in printed and electronic form:• Lecture Notes in Computer Science (LNCS), incl. its subseries Lecture Notes in Artificial Intelligence (LNAI) and Lecture Notes in Bioinformatics (LNBI), and LNCS Transactions; • Lecture Notes in Business Information Processing (LNBIP);• Communications in Computer and Information Science (CCIS);• Lecture Notes of the Institute for Computer Sciences, Social Informatics and Tele-communications Engineering (LNICST);• IFIP Advances in Information and Communication Technology (IFIP AICT), for-merly known as the IFIP Series;• Proceedings in Information and Communications Technology (PICT).Your contribution may be prepared in LaTeX or Microsoft Word. Technical Instruc-tions for working with Springer’s style files and templates are provided in separate documents which can be found in the respective zip packages on our website. * No academic titles or descriptions of academic positions should be included in the addresses. Either this information should be omitted altogether (preferably), or it should be included in a footnote at the end of the first page. Information of this nature, given in the addresses, will be deleted by our typesetters.22Preparation of Your Paper2.1Structuring Your PaperAffiliations. The affiliated institutions are to be listed directly below the names of the authors. Multiple affiliatons should be marked with superscript arabic numbers, and they should each start on a new line as shown in this document. In addition to the name of your affiliation, we would ask you to give the town and the country in which it is situated. If you prefer to include the entire postal address, then please feel free to do so. E-mail addresses should start on a new line and should be grouped per affiliation. Headings.Headings should be capitalized (i.e., nouns, verbs, and all other words except articles, prepositions, and conjunctions should be set with an initial capital) and should, with the exception of the title, be aligned to the left. Only the first two levels of section headings should be numbered, as shown in Table 1. The respective font sizes are also given in Table 1. Kindly refrain from using “0” when numbering your section headings.Table 1. Font sizes of headings. Table captions should always be positioned above the tables. Heading level Example Font size and style Title (centered) Lecture Notes14 point, bold1st-level heading 1 Introduction12 point, bold2nd-level heading 2.1 Printing Area10 point, bold3rd-level heading Run-in Heading in Bold. Text follows10 point, bold4th-level heading Lowest Level Heading. Text follows10 point, italic Words joined by a hyphen are subject to a special rule. If the first word can stand alone, the second word should be capitalized.Here are some examples of headings: “Criteria to Disprove Context-Freeness of Collage Languages”, “On Correcting the Intrusion of Tracing Non-deterministic Pro-grams by Software”, “A User-Friendly and Extendable Data Distribution System”, “Multi-flip Networks: Parallelizing GenSAT”, “Self-determinations of Man”. Lemmas, Propositions, and Theorems.The numbers accorded to lemmas, proposi-tions, and theorems, etc. should appear in consecutive order, starting with Lemma 1. Please do not include section counters in the numbering like “Theorem 1.1”.2.2Page Numbering and Running HeadsThere is no need to include page numbers or running heads; this will be done at our end. If your paper title is too long to serve as a running head, it will be shortened. Your suggestion as to how to shorten it would be most welcome.32.3 Figures and TablesIt is essential that all illustrations are as clear and as legible as possible. Vector graph-ics – instead of rasterized images – should be used for diagrams and schemas when-ever possible. Please check that the lines in line drawings are not interrupted and have a constant width. Grids and details within the figures must be clearly legible and may not be written one on top of the other. Line drawings are to have a resolution of at least 800 dpi (preferably 1200 dpi). The lettering in figures should not use font sizesp o w e r f l u c t u a t i o n (1555 n m ), d B time, msFig. 1. Power distribution of channel at 1555 nm along the link of 383 km (Source: LNCS 5412, p. 323)Fig. 2. Artifacts empowered by Artificial Intelligence (Source: LNCS 5640, p. 115)4smaller than 6 pt (~ 2mm character height). Figures are to be numbered and to have a caption which should always be positioned under the figures, in contrast to the caption belonging to a table, which should always appear above the table.Captions are set in 9-point type. If they are short, they are centered between the margins. Longer captions, covering more than one line, are justified (Fig. 1and Fig. 2 show examples). Captions that do not constitute a full sentence, do not have a period.Text fragments of fewer than four lines should not appear at the tops or bottoms of pages, following a table or figure. In such cases, it is better to set the figures right at the top or right at the bottom of the page. A figure should never be placed in the mid-dle of a paragraph.If screenshots are necessary, please make sure that the essential content is clear to the reader.Remark 1.In the printed volumes, illustrations are generally black and white (half-tones), and only in exceptional cases, and if the author or the conference organization is prepared to cover the extra costs involved, are colored pictures accepted. Colored pictures are welcome in the electronic version free of charge. If you send colored figures that are to be printed in black and white, please make sure that they really are also legible in black and white. Some colors show up very poorly when printed in black and white.2.4FormulasDisplayed equations or formulas are centered and set on a separate line (with an extra line or half line space above and below). Displayed expressions should be numbered for reference. The numbers should be consecutive within the contribution, with num-bers enclosed in parentheses and set on the right margin. Please do not include section counters in the numbering.x + y = z(1) Equations should be punctuated in the same way as ordinary text but with a small space before the end punctuation mark.2.5FootnotesThe superscript numeral used to refer to a footnote appears in the text either directly after the word to be discussed or – in relation to a phrase or a sentence – following the punctuation mark (comma, semicolon, or period).1For remarks pertaining to the title or the authors’ names, in the header of a paper, symbols should be used instead of a number (see first page of this document). Please note that no footnotes may be included in the abstract.1The footnote numeral is set flush left and the text follows with the usual word spacing.5 2.6Program CodeProgram listings or program commands in the text are normally set in typewriter font: program Inflation (Output){Assuming annual inflation rates of 7%, 8%, and10%,... years};const MaxYears = 10;var Year: 0..MaxYears;Factor1, Factor2, Factor3: Real;beginYear := 0;Factor1 := 1.0; Factor2 := 1.0; Factor3 := 1.0;WriteLn('Year 7% 8% 10%'); WriteLn;repeatYear := Year + 1;Factor1 := Factor1 * 1.07;Factor2 := Factor2 * 1.08;Factor3 := Factor3 * 1.10;WriteLn(Year:5,Factor1:7:3,Factor2:7:3,Factor3:7:3)until Year = MaxYearsend.[Example of a computer program from Jensen K., Wirth N.: Pascal User Manual and Report. Springer, New York (1991)]2.7Citations and BibliographyFor citations in the text, we prefer the use of square brackets and consecutive num-bers. Numbers should be grouped where appropriate. We would write [1-5] but [1], [3], [5], etc. The numbers in the bibliography section are without square brackets. Citations using labels or the author/year convention are also acceptable.Please write all references using the latin alphabet. If the title of the book you are referring to is, e.g., in Russian or Chinese, then please write (in Russian) or (in Chi-nese) at the end of the transcript or translation of the title.In order to permit cross referencing within , and even-tually between different publishers and their online databases, Springer standardizes the format of the references. This feature aims to increase the visibility of publications and facilitate academic research. Please base your references on the examples given in the references section of these instructions. References that do not adhere to this style will be reformatted at our end.We would like to draw your attention to the fact that references to LNCS proceed-ings papers are particularly often reformatted due to missing editor names or incom-plete publisher information. This adjustment may result in the final papers as pub-lished by Springer having more pages than the original versions as submitted by the authors. Here is an example:6Reference as formatted in author’s original version:Assemlal, H.E., Tschumperlé, D., Brun, L.: Efficient Computation of PDF-Based Characteris-tics from Diffusion MR Signal. In: MICCAI. Volume 5242. (2008) 70–78Reference after reformatting by Springer:Assemlal, H.E., Tschumperlé, D., Brun, L.: Efficient Computation of PDF-Based Characteris-tics from Diffusion MR Signal. In: Metaxas, D., Axel, L., Fichtinger, G., Székely, G. (eds.) MICCAI 2008, Part II. LNCS, vol. 5242, pp. 70–78. Springer, Heidelberg (2008)One more line is needed for this reference, as a result of Springer’s adjustment.Please make sure that all your sources are correctly listed in the reference section. Do not include references to pieces of work that are not connected with your paper.The references section at the end of this paper shows a sample reference list with entries for journal articles [1], an LNCS chapter [2], a book [3], proceedings without editors [4] and [5], as well as a URL [6]. Please note that Springer proceedings are cited with their publication acronyms and volume numbers.2.8PlagiarismSpringer takes plagiarism seriously. If an author has copied from another author or has used parts of another author’s work, without his or her permission and a refer-ence, then the paper on SpringerLink will be given a “retracted” stamp, and an erra-tum explaining the reasons for the retraction will be included. In addition, the volume editors and the author’s academic supervisors will be informed. Acknowledgements. This should always be a run-in heading and not a section or subsection heading. It should not be assigned a number. The acknowledgements may include reference to grants or supports received in relation to the work presented in the paper.3Additional Information Required from Authors3.1Copyright FormThere are different copyright forms in place for the different Springer Computer Science proceedings book series. A prefilled copyright form is usually available from the conference website. Alternatively, the form may be downloaded from the web-page of the respective series on . Please send your signed copyright form to your conference publication contact, either as a scanned PDF or by fax or by courier. One author may sign on behalf of all of the other authors of a par-ticular paper. In this case, the author signs for and accepts responsibility for releasing this material on behalf of any and all co-authors. Digital signatures are also accepta-ble.73.2Contact Author InformationKindly assure that, when you submit the final version of your paper, you also provide the name and e-mail address of the contact author for your paper. These details are used to compile a list of contact authors for our typesetting partners SPS in India. The contact author must be available to check the paper roughly seven weeks before the start of the conference, or before the book is due to leave the printing office, in the case of post-conference proceedings. He or she will be contacted by SPS from the following e-mail address: typesetting@sps.co.in.3.3Correct Representation of Author NamesAuthors’ names should be written out in full at the tops of the papers. They are short-ened by us to “initials surname” in the running heads and will take the form “sur-name, given name” in the author index. If you or any of your co-authors have more than one family name, it should be made quite clear how your name is to be displayed in the running heads and the author index. Chinese authors should write their given names in front of their surnames at the tops of their papers. If you only have one (main) name, please make sure that this name is written out in full in the running heads, when you check your final PDF.4Typesetting of Your Paper at SpringerPlease make sure that the paper you submit is final and complete, that any copyright issues have been resolved, that the authors listed at the top of the chapter really are the final authors, and that you have not omitted any references. Following publication, it is not possible to alter your paper on SpringerLink. Kindly note that we prefer the use of American English.4.1What Will Be Done with Your PaperIf the instructions have been followed closely, then only very minor alterations will be made to your paper. The production team at SPS checks the format of the paper, and if, for example, vertical spacing has been inserted or removed, making the paper unsightly or difficult to read, then this is remedied. In addition, running-heads, final page numbers and a copyright line are inserted, and the capitalization of the headings is checked and corrected if need be. Finally, the reference section is attuned to our specifications (see also Section 2.7).4.2Proof Reading StageOnce the files have been worked upon, SPS sends a copy of the final PDF of each paper to its contact author. The contact author is asked to check through the final PDF8to make sure that no errors have crept in during the transfer or preparation of the files. This should not be seen as an opportunity to update or copyedit the paper, which is not possible due to time constraints. Only errors introduced during the preparation of the files will be corrected. Particular attention should be paid to the references section.If SPS does not receive a reply from a particular contact author, within the time-frame given (usually 72 hours), then it is presumed that the author has found no errors in the paper. The tight publication schedule of our proceedings series does not allow SPS to send reminders or search for alternative e-mail addresses on the Internet.In some cases, it is the contact volume editor or the publication chair who checks all of the PDFs. In such cases, the authors are not involved in the checking phase.The purpose of the proof is to check for typesetting or conversion errors and the completeness and accuracy of the text, tables and figures. Substantial changes in con-tent, e.g., new results, corrected values, title and authorship, are not possible and can-not be processed.5Online Publication in SpringerLinkAll papers will be published in our digital library . Only subscribers to Springer’s eBook packages or to the electronic book series are able to access the full text PDFs and references of our online publications. Meta-data, ab-stracts and author e-mail addresses are freely available for all users.6Checklist of Items to Be Sent to Volume Editor•The final source files, incl. any non-standard fonts.• A final PDF file corresponding exactly to the final source files.• A copyright form, signed by one author on behalf of all of the authors of the paper. •The name and e-mail address of the contact author who will check the proof of the paper.• A suggestion for an abbreviated running head, if appropriate.•Information about correct representation of authors’ names, where necessary. References1.Smith, T.F., Waterman, M.S.: Identification of Common Molecular Subsequences. J. Mol.Biol. 147, 195–197 (1981)2.May, P., Ehrlich, H.C., Steinke, T.: ZIB Structure Prediction Pipeline: Composing a Com-plex Biological Workflow through Web Services. In: Nagel, W.E., Walter, W.V., Lehner, W. (eds.) Euro-Par 2006. LNCS, vol. 4128, pp. 1148–1158. Springer, Heidelberg (2006) 3.Foster, I., Kesselman, C.: The Grid: Blueprint for a New Computing Infrastructure. Mor-gan Kaufmann, San Francisco (1999)4.Czajkowski, K., Fitzgerald, S., Foster, I., Kesselman, C.: Grid Information Services forDistributed Resource Sharing. In: 10th IEEE International Symposium on High Perfor-mance Distributed Computing, pp. 181–184. IEEE Press, New York (2001)9 5.Foster, I., Kesselman, C., Nick, J., Tuecke, S.: The Physiology of the Grid: an Open GridServices Architecture for Distributed Systems Integration. Technical report, Global Grid Forum (2002)6.National Center for Biotechnology Information, Appendix: Springer Author DiscountAuthors contributing to any of Springer’s computer science proceedings publications are entitled to a 33.3% discount off all Springer products when placing an order through . Please go to /authors, then click on ‘Book Authors’ and then follow the instructions given in the Book Dis-counts section. You are requested to give full details of your Springer publication to obtain a so-called SpringerToken. This token is a number that must be entered when placing an order through our online catalogue, in order to obtain the discount. Contact UsIf you have any further questions regarding the preparation of your paper, then please do not hesitate to get in touch with us.•For all questions related to our LaTeX style files, your contact person is: Mr. Frank Holzwarth, e-mail: frank.holzwarth@.•For overall technical questions concerning the preparation of LNCS/LNAI/LNBI papers, please contact Ms. Anna Kramer, e-mail: lncs@.•For the LNBIP series, please contact Ms. Viktoria Meyer, e-mail: lnbip@.•For the CCIS series, please contact Ms. Leonie Kunz, e-mail:ccis@.•For the LNICST series, please contact Mr. Peter Strasser, e-mail:lnicst@.•For the IFIP AICT series, please contact Ms. Erika Siebert-Cole, e-mail: ifip@.•For the PICT series, please contact Mr. Masayuki Nakamura, email: pict@springer.jp.[April 19, 2011]。
Managing the Complexity of Performance Monitoring Hardware:The Brink and Abyss ApproachBrinkley SpruntElectrical Engineering DepartmentBucknell UniversitySubmitted to the International Journal of High Performance Computing Applications for review,May2004.Suggested subject areas for reviews:-implementation techniques and pragmatic software andarchitectural considerations-performance studiesProposed running head:Managing Performance Counter ComplexityAuthor Contact Information:Brinkley SpruntElectrical Engineering DepartmentBucknell UniversityMoore AvenueLewisburg,PA17837(570)577-3397(office)(570)577-1234(department office)(570)577-1449(department FAX)Email:bsprunt@Home Page:/~bsprunt/Author Biography:Brinkley Sprunt is an assistant professor of Electrical Engineeringat Bucknell University.His research interests include computerperformance modeling,measurement,and optimization.Prior tojoining Bucknell,he worked at Intel Corporation for nine yearswhere he was a member of the architecture teams for the80960,Pentium Pro,and Pentium4projects.On the Pentium4project,heled the performance validation team and was the principle architectof the Pentium4performance monitoring capabilities.1SummaryRecent microprocessor advances have significantly improved the capabilities of on-chip per-formance monitoring hardware,as is evident in the Pentium4processor[12].However,the com-plexity of the performance monitoring hardware has also risen,making the tasks of understanding and using the performance monitoring hardware much more difficult for performance analysts and software developers.Additionally,documentation for these features is becoming increasingly cryp-tic[6],further exacerbating the task of using these features effectively.This paper describes a new approach for managing this complexity that enables the full feature set of this hardware to be used while still providing a high-level,easy-to-use user interface.This approach augments reference manual documentation by creating XML[15]descriptions of the performance monitoring hardware capabilities and its desired configuration.This XML-based approach was used to develop the brink and abyss tools[11],which provide high-level interface to the Pentium4performance monitoring capabilities.2IntroductionIntel’s introduction of the Pentium4family of processors brought a number of computer de-sign advances to a microprocessor for thefirst time.Among these advances are the inclusion of a trace cache,low-latency double-pumped ALUs,a deep pipeline,and support for simultane-ous multithreading[5,9].Along with these performance features,the Pentium4designers also significantly improved the Pentium4’s performance monitoring hardware(typically referred to as EMON,or event-monitoring hardware)compared to previous Pentium-family processors.These EMON improvements include support for many more performance event detectors and counters, thread-based event detection and counting,the ability to obtain non-speculative as well as specu-lative event counts,and support for precise event-based sampling[12].However,these advanced EMON capabilities along with the aggressive processor design combine to make the task of un-derstanding and using these performance monitoring features significantly more difficult than with previous processors.This paper describes a new approach for managing the complexities associated with these EMON advances that provides access to the the full feature set of the hardware while still providing a high-level,easy-to-use interface to the user.This approach was used to develop the brink and abyss tools[11],which provide a high-level interface to the EMON capabilities of the Pentium4 on Linux systems.The brink and abyss tools rely upon an Extensible Markup Language(XML) [15]description of the performance monitoring capabilities of the Pentium4along with an XML description of the desired EMON configuration and the applications to be monitored.These XML descriptions provide significant advantages to the users as well as the maintainers of these tools.The XML description of the Pentium4EMON capabilities,referred to as the EMON configu-rationfile,encapsulates the intricacies of the Pentium4EMON hardware while still being human-readable and easy to maintain.Many details of the Pentium4EMON hardware must be describedin the XML EMON configurationfile,including the large number of event detectors and counters, the subset of all events that are detectable by each event detector,the subset of all counters that can count events detected by each event detector,and the many bitfields that are used to enable and configure the various event detection and qualification options.The XML representation allows all of this information to be described in a human-readable form while simultaneously defining the basic data structures needed by software tools to manipulate the performance monitoring hard-ware.Additionally,the EMON configurationfile defines all of the base performance events that can be monitored along with the various event tagging and sampling mechanisms.This XML-based approach for describing these capabilities and mechanisms significantly improves the documenta-tion of these features compared to a mere description of the performance monitoring capabilities and events in a reference manual,as is the current practice.Once this information is presented and maintained in XML form,updating the EMON software tools as new events,counters,and/or features are added to successive versions of a processor requires only simple edits to the XML description and little,if any,code development.The XML description of the desired performance monitoring configuration and the applications to be monitored,referred to as the experimentfile,also provides significant benefits to the mi-croprocessor performance analyst.The large number of events and event qualification options available in the Pentium4result in a myriad of combinations which make the creation of pre-defined event names infeasible.As such,the XML-based experimentfile allows the user to define names for each event to be monitored,providing greatflexibility.Also,the configuration of the performance events to be monitored is done symbolically,using symbols defined in the EMON configurationfile.This eliminates the need for the user to remember or reference documentation describing the ranges for configuration register bitfields.The event descriptions in the experiment file are customizations of the base events defined in the EMON configurationfile.The experi-mentfile also does not require the user to allocate event detectors and/or counters for the desired performance events.This allocation is handled completely by the brink and abyss tools,further simplifying the configuration task.T o ease the collection of multiple performance events across multiple applications,the experimentfile allows multiple application programs and multiple perfor-mance events sets per application run to be specified.In this manner,the user can define a set of performance events and collect the corresponding EMON data across multiple applications in one fell swoop.The outline of this paper is as follows.First,a general description of EMON capabilities along with brief descriptions of other EMON software tools is given.This discussion also highlights recent advances in performance monitoring techniques and the rising complexity of EMON hard-ware.Second,the Pentium4performance monitoring hardware interface is described.Third,the brink and abyss tools are described in detail including the goals for these tools,the brink user interface,the abyss front end,and the low-level abyss device driver that directly manipulates the Pentium4EMON configuration.Fourth,specific examples are discussed of brink and abyss us-age along with brief descriptions of example performance data.Fifth,a demonstration of how the XML EMON configurationfile can easily be updated for successive models of the Pentium4is given.This paper concludes with a summary of the advantages of the XML-based approach used to create the brink and abyss tools.3BackgroundMost modern,high-performance processors have on-chip hardware,typically referred to as EMON(event monitoring)hardware,that can be used to monitor the performance of the processor [13].EMON data can be used to understand how applications,the operating system,and the processor are performing.These data can then be used to guide efforts to improve performanceby tuning the algorithms used by the application and/or the operating system and by tuning the code sequences used to implement those algorithms.These performance data can also be used to guide the design of new processor implementations.EMON hardware is used to capture counts of various performance events that occur during the execution of applications and the operating system.These performance events typically fall into one offive categories:program characterization,memory,pipeline stalls,branch prediction, and resource utilization.The program characterization events are used to characterize the at-tributes of a program(and/or the operating system)that are largely independent of the processor’s implementation,such as loads,stores,and branches.The memory events are used to analyze memory hierarchy performance and include events such as cache misses and bus transactions. The pipeline stall events are used to analyze how well the program’s instructionsflow through the processor’s pipeline.The branch prediction events allow one to analyze the performance of the branch prediction hardware that is used predict the correct instruction execution path.The re-source utilization events allow one to monitor how often certain resources in a processor are used (e.g.,the number of cycles an instruction issue port is being used).3.1Basic Hardware Performance Monitoring:Event Detectors and CountersEMON hardware is typically composed of two components:performance event detectors and event counters.By properly configuring the event detectors and counters,one can obtain counts of a variety of performance events under various conditions.Performance event detectors can be configured to detect any one of a large number of perfor-mance events such as cache misses or page faults.Event detectors typically have an event mask field that allows further qualification of the event(such as detecting the state of an accessed cache line as being in the modified,shared,exclusive,or invalid states).Event detectors also providethe ability to qualify event detection by the current privilege level of the processor,allowing events caused by the operating system and/or user-level programs to be selected.Event counters are used to count the events detected by the performance event detectors.The counters can also be configured to count only under certain edge and threshold conditions.The edge detection feature allows one to configure the counter to count only transitions between the absence of an event and the presence of an event(or vice-versa),which is helpful to determine the duration of events such as pipeline stalls.A generalization of the edge detection feature allows the counter to be incremented only when the number of reported events during a cycle is less than or greater than a specified threshold.For example,one can use the threshold detection support to count the number of times a superscalar processor completed more than the specified threshold number of instructions in one cycle.3.2Event-Based SamplingAlthough performance event detectors and counters can easily identify the presence of a per-formance problem(e.g.,a high cache miss rate or a high branch misprediction rate),it is usually necessary tofind the code locations(in the application and/or the operating system)that are causing the the performance problem.Once the programmer knows the source of the perfor-mance problem,the high-level algorithms used by the application and/or the low-level code can be altered to avoid or reduce the impact of the performance problem.A common technique used tofind these problematic code locations is to use event-based sampling(EBS).When conducting EBS,the EMON hardware is configured to cause an interrupt after N performance events have been observed.The interrupt service routine(ISR)collects sample data about the eventing instruction,such as the program counter(PC)for the eventing instruction.By creating a histogram of the sample PC values,one can identify the code sectionscausing the majority of the performance problems in question.Although EBS is an effective approach for identifying the source of performance problems,the EBS support provided by most processors is imprecise.The imprecision arises because it is often the case that last instruction retired prior to the occurrence EBS interrupt(and the source of the sample data)is not the instruction that caused the performance event.The processor’s pipeline and the delay between detecting the Nth performance event and the signaling of an interrupt com-bine to create an arbitrary distance(in completed instructions)between the eventing instruction and the PC sample saved by the EBS ISR.These inaccuracies often make it difficult to use and interpret the EBS data properly.3.3Performance Monitoring Tools and CapabilitiesA variety of software tools are available that provide a high-level interface and analysis capa-bilities for performance monitoring hardware.The P API tool suite[2]and the PCL library[1]both provide common interfaces to performance monitoring hardware for many different processors.In-tel’s VT une Performance Analyzer[7]supports the performance monitoring hardware of all of the Intel Pentium and Itanium processors and provides a number of additional performance analysis tools such as call graph profiling and processor-specific tuning advice.The Compaq(Digital)Con-tinuous Profiling Infrastructure for Alpha processors[3]uses performance monitoring hardware to provide a low-overhead facility for monitoring the performance of an entire system.The Rabbit Performance Counter Library[4]provides a high-level interface to Intel P6-based Pentium proces-sors and AMD Athlon processors on Linux systems.The brink and abyss tools[11],described in this paper,provide a high-level interface to the Pentium4performance monitoring features on a Linux system.3.4Recent Advances in Performance Monitoring CapabilitiesIntel’s introduction of the Pentium4marks a significant increase in the general capabilities found in microprocessor performance monitoring hardware[12].The most noticeable advance is the inclusion of significantly more event detectors and counters than in previous processors. Before Pentium4,microprocessors with EMON support typically included only two to six event detectors and counters.The Pentium4supports48event detectors and18event counters.This larger number of event counters enables much more performance data to be collected simultane-ously than on previous processors,reducing the number of application runs to collect the desired data and reducing the inaccuracies introduced by multiple application runs.The Pentium4also introduced instruction tagging mechanisms that allow nonspeculative per-formance event counts to be obtained.On speculative-execution processors,such as the Pentium 4,most instructions are executed before they are known to be on the correct execution path.With-out a mechanism to separate the detected performance events into those caused by speculatively-executed instructions that never complete from instructions that do complete(retire),erroneous conclusions about program behavior can be drawn.The Pentium4’s instruction tagging mecha-nisms operate as follows.As the Pentium4decodes each instruction,it breaks it into a sequence of one or more simple operations called micro-operations or uops.The Pentium4tagging mecha-nisms enable these uops to be tagged when they cause certain performance events.Once a uop is tagged,it retains this tag until it retires successfully or is cancelled.As the uops pass through the retirement logic,tagged uops that retire can be counted,providing a nonspeculative event count.The Pentium4also uses its uop tagging mechanism along with microcode-level traps rather than macroinstruction-level interrupts to provide support for precise event-based sampling(PEBS). By recognizing tagged uops as they are about to retire and using a microcode-level trap to collectEBS data,the imprecision of prior EBS implementations is avoided because no uops(or instruc-tions)are allowed to retire between the retirement of the eventing instruction and the capturing of the sample data.The Pentium4also provides a special sample buffering mechanism that auto-matically collects many samples before a macro-level interrupt service routine must be invoked to save the samples.This buffering technique significantly lowers the overhead associated with EBS, resulting in fewer perturbations of the application being monitored.The Pentium4is also thefirst microprocessor to support simultaneous multithreading.Simul-taneous multithreading(SMT)is a technique that allows more than one task(thread)to share the processor’s resources concurrently[14,8].Thus,a single SMT processor appears to the operat-ing system as multiple processors and the operating system will execute ready-to-run threads for each processor it sees.The SMT approach provides a throughput advantage over a non-threaded processor by allowing the processor’s resources to continue to be used when one execution thread becomes stalled(e.g.,due to a cache miss).The Pentium4processor provides SMT support by allowing its resources to be shared by two threads in dual-thread mode.The Pentium4can also operate in single-thread mode where the whole processor is allocated to only one thread.The Pentium4EMON hardware supports these SMT features by allowing event detection to be qualified by thread ID and thread mode.Each Pentium4event detector allows one to specify that event detection should only occur for thread0,or thread1,or for either thread.Each Pentium 4event counter allows one to specify that event counting should occur when operating in single-thread mode,dual-thread mode,any-thread mode,or no-thread mode.In no-thread mode,neither thread is active,but the processor may still respond to bus transactions initiated by other active processors in the system.3.5The Complexity of Performance Monitoring HardwareThe recent advances in performance monitoring capabilities discussed above significantly in-crease the complexity of the EMON configuration for a desired performance experiment.This increase in configuration complexity arises from several factors.First,the increased number of event detectors and counters along with the restriction that specific events can only be detected and counted by subsets of the event detectors and counters makes it difficult to allocate event detectors and counters,especially when data for more than just a few events need to be collected simultaneously.Second,the Pentium4supports many qualification options for event detection and event counting.For example,the detection and counting of most events can be qualified by thread ID,privilege level,thread mode,and threshold level.In addition to these generic qualification op-tions,each event supports the use of a16-bit event-specific mask that can be used to further qual-ify an event(e.g.,selecting among taken/not-taken or predicted/mispredicted branches).Third,the Pentium4instruction tagging mechanisms require multiple events to be configured:the event that causes the instruction to be tagged and the event that detects tagged instructions as they retire. Furthermore,some of the instruction tagging mechanisms require the configuration of ancillary control registers associated with the tagging mechanisms.Fourth,the Pentium4PEBS support requires the configuration of tagged events,the configuration of the sample buffering mechanism, and the periodic“harvesting”of the buffered samples.When these EMON capabilities are com-pared to the EMON capabilities of prior processors,that only supported a few counters and event detectors and had no thread or PEBS support,one can easily see that the power of the Pentium 4EMON capabilities comes at the price of increased configuration complexity.4The Pentium4EMON InterfaceThe interface between system software and Pentium4EMON hardware is complex.Even the simplest configuration to detect and count one event requires the setup of two EMON machine-specific registers(MSRs),an event detector and an event counter.An EMON configuration for more than one event requires the careful allocation of event detectors and event counters,as only a subset of the48event detectors and18counters can be used to count a particular event.This section provides a brief overview of the the Pentium4EMON interface,discussing the organization of the event detectors and counters along with their configuration,the use of the uop-tagging mechanisms for obtaining non-speculative event counts,and the PEBS configuration and event sample buffering mechanism.4.1Event Detectors and CountersThe basic elements of the Pentium4EMON hardware are the event detectors and event coun-ters.These are configured via to types of MSRs,the Event Select Control Register(ESCR)and the Counter Configuration Control Registers(CCCR).An ESCR is used to select the event to be detected and the CCCR is used to configure the counter for the detected events.Each CCCR is also paired with a40-bit counter that holds the count for the detected events.As mentioned previ-ously,an ESCR can only detect a subset of the available EMON events and the detected events from an ESCR can only be counted on a subset of the available CCCR/counter pairs.Figure1 shows the available interconnections between the ESCRs and CCCR/counter pairs.Each grey box on the left side of Figure1represents a unit on the Pentium4processor that contains event detectors.T o support the simultaneous counting of the same event from two different threads while running dual-thread mode,each of these units has two similarly-equipped ESCRs.The units containing ESCRs are grouped into four sets shown horizontally in Figure1.Each of these ESCRFigure1:Pentium4ESCR and CCCR/Counter Interconnections.sets shares one group of CCCR/counter pairs,shown on the right side of Figure1.The names of the Pentium4units that contain the ESCRs and CCCR/counter pairs are noted in the top left of each box in the Figure1.These names are based upon the function of the Pentium4unit.For example,the Instruction Queue unit(IQ)hosts two event detectors associated with the instruction queuing and scheduling events and also hosts a group of CCCR/counter pairs that is shared by the IQ,ALF,RA T,and CRU units.The ESCR select values indicated in the lower left of each unit box are used when configuring a CCCR to indicate which ESCR’s events should be counted.Each ESCR supports the selection of a particular event and allows the detection of that event to be qualified by an event-specific mask,thread ID,and privilege level.The layout of the ESCRs is shown in Figure2.The ESCR event mask allows subsets of the event selected to be counted.For example,the instrUSR,T1USR, and T0USR bit should be set.The tag enable and tag valuefields of the ESCR are used by the execution tagging mechanism discussed in section4.2.The CCCR(see Figure3)allows event counting to be qualified by thread mode and thresh-old and also supports several features related to counter overflow.Since the Pentium4supports simultaneous multithreading(SMT),the processor may execute in one of several SMT modes: single-thread mode,dual-thread mode,and no-thread mode.The CCCR supports qualification of event counting for each or any of these modes(single-thread,dual-thread,no-thread,or any-thread mode).The CCCR also allows event counting to be qualified by a threshold.The threshold support allows counting to occur when the number of events reported in one cycle from an event detector is greater-than or less-than-or-equal to the specified threshold.The compare,comple-ment,edge,and threshold bitfields in the CCCR are used to configure the threshold support.The CCCR also contains an overflowflag bit(OVF)that is set when the counter overflows.The CCCR can be configured to request an interrupt on overflow(via the OVF T0and OVF T1bits) and also be forced to overflow on every event occurrence(via the Force3210Figure2:Pentium4Event Select Control Register(ESCR).4.2Tagging Mechanisms for Obtaining Non-speculative Event CountsThe Pentium4tagging mechanisms allow non-speculative event counts to be obtained by enabling uops that encounter certain events to be tagged and counting tagged uops as they retire. Three different tagging mechanisms are implemented by the Pentium4:front-end,execution,and replay.The front-end tagging mechanism is able to tag uops responsible for events that occur early in the pipeline that are related to instruction fetching,instruction types,and uop delivery from the trace cache.The execution tagging mechanism is able to tag certain arithmetic uops as they write their results back to the registerfile.This mechanism uses a user-specified,4-bitfield to tag the uops(via the tag enable and the tag valuefields in the ESCR).These uops can be tested as they retire and counted if any of these tag bits are set.The replay tagging mechanism is able to tag uops that are re-issued(replayed)due to various conditions such as cache misses,branch mispredictions,dependence violations,and resource conflicts.The replay tagging mechanism relies upon two other MSRs that must be configured to select the desired replay cause(e.g.,a cache miss)as well as the desired uop type(a load).3210Figure3:Pentium4Counter Configuration Control Register(CCCR).4.3Event-Based Sampling SupportThe Pentium4support for precise event-based sampling(PEBS)provides two main advan-tages over prior processors:accuracy and low sampling overhead.Prior processors that relied solely upon an interrupt service routine to collect sample data upon performance counter overflow were inaccurate because often the data collected did not correspond to the instruction that caused the performance counter to overflow.Furthermore,an interrupt was needed to collect each sam-ple and frequent interrupts can significantly perturb the behavior of the system being monitored. The Pentium4overcomes these problems by using a microcode assist to collect sample data for tagged instructions as they retire and by buffering the samples collected.The microcode assist always collects the sample data from a tagged(an event-causing)instruction,so the information is precise.Each sample collected is placed directly in a memory buffer by the microcode assist and only when the buffer is approaching capacity is an interrupt service routine invoked to harvest the samples.The setup of the Pentium4PEBS support is complex.In addition to configuring the tagging mechanism for the desired EMON event,the Pentium4debug store(DS)area must be configured and a PEBS interrupt service routine must be installed.The DS configuration specifies the base, threshold,and maximum memory addresses for the PEBS buffer along with the buffer index and the reset value for the event counter.The PEBS interrupt service routine harvests the collected samples and re-enable the PEBS interrupt.5The Brink and Abyss ToolsThe overall goal for the brink and abyss tools[11]is to provide a high-level interface to the EMON capabilities of the Pentium4on Linux systems.This high-level interface allows the user to concentrate upon the types of performance data to be collected while ignoring many of the complexities of the Pentium4EMON hardware and its configuration.However,this goal is not unlike the goals for other tools that support performance counter hardware such as P API[2], PCL[1],VT une[7],DCPI[3],and Rabbit[4].The key difference in between the brink and abyss tools and these other tools is the XML-based approach used by brink and abyss to document and describe the Pentium4’s EMON events and capabilities as well as the user’s desired EMON configuration.This XML-based approach fulfills several secondary goals of the brink and abyss tools,namely:to improve the documentation of the Pentium4EMON capabilities,to provide enoughflexibility to enable all the EMON features to be used,and to ease the maintenance of these tools as new Pentium4implementations add new events and features.This section describes the implementation of the brink and abyss tools and provides examples of their use.First,the overall structure of the tools is described.Next,the brink user interface along with the Pentium4EMON configuration and experimentfiles are described.The abyss front end and device driver are then discussed.This section closes with usage and maintenance examples。
