Debugging multi-agent systems using design artifacts The case of interaction protocols

计算机类学术期刊和会议影响因子Estimated impact of publication venues in Computer Science (higher is better)1. OSDI: 3.31 (top 0.08%)2. USENIX Symposium on Internet Technologies and Systems:3.23 (top 0.16%)3. PLDI: 2.89 (top 0.24%)4. SIGCOMM: 2.79 (top 0.32%)5. MOBICOM: 2.76 (top 0.40%)6. ASPLOS: 2.70 (top 0.49%)7. USENIX Annual Technical Conference: 2.64 (top 0.57%)8. TOCS: 2.56 (top 0.65%)9. SIGGRAPH: 2.53 (top 0.73%)10. JAIR: 2.45 (top 0.81%)11. SOSP: 2.41 (top 0.90%)12. MICRO: 2.31 (top 0.98%)13. POPL: 2.26 (top 1.06%)14. PPOPP: 2.22 (top 1.14%)15. Machine Learning: 2.20 (top 1.22%)16. 25 Years ISCA: Retrospectives and Reprints: 2.19 (top 1.31%)17. WWW8 / Computer Networks: 2.17 (top 1.39%)18. Computational Linguistics: 2.16 (top 1.47%)19. JSSPP: 2.15 (top 1.55%)20. VVS: 2.14 (top 1.63%)21. FPCA: 2.12 (top 1.71%)22. LISP and Functional Programming: 2.12 (top 1.80%)23. ICML: 2.12 (top 1.88%)24. Data Mining and Knowledge Discovery: 2.08 (top 1.96%)25. SI3D: 2.06 (top 2.04%)26. ICSE - Future of SE Track: 2.05 (top 2.12%)27. IEEE/ACM Transactions on Networking: 2.05 (top 2.21%)28. OOPSLA/ECOOP: 2.05 (top 2.29%)29. WWW9 / Computer Networks: 2.02 (top 2.37%)30. Workshop on Workstation Operating Systems: 2.01 (top 2.45%)31. Journal of Computer Security: 2.00 (top 2.53%)32. TOSEM: 1.99 (top 2.62%)33. Workshop on Parallel and Distributed Debugging: 1.99 (top 2.70%)34. Workshop on Hot Topics in Operating Systems: 1.99 (top 2.78%)35. WebDB (Informal Proceedings): 1.99 (top 2.86%)36. WWW5 / Computer Networks: 1.97 (top 2.94%)37. Journal of Cryptology: 1.97 (top 3.03%)38. CSFW: 1.96 (top 3.11%)39. ECOOP: 1.95 (top 3.19%)40. Evolutionary Computation: 1.94 (top 3.27%)41. TOPLAS: 1.92 (top 3.35%)42. SIGSOFT FSE: 1.88 (top 3.43%)43. CA V: 1.88 (top 3.52%)44. AAAI/IAAI, Vol. 1: 1.87 (top 3.60%)45. PODS: 1.86 (top 3.68%)46. Artificial Intelligence: 1.85 (top 3.76%)47. NOSSDA V: 1.85 (top 3.84%)48. OOPSLA: 1.84 (top 3.93%)49. ACM Conference on Computer and Communications Security: 1.82 (top 4.01%)50. IJCAI (1): 1.82 (top 4.09%)51. VLDB Journal: 1.81 (top 4.17%)52. TODS: 1.81 (top 4.25%)53. USENIX Winter: 1.80 (top 4.34%)54. HPCA: 1.79 (top 4.42%)55. LICS: 1.79 (top 4.50%)56. JLP: 1.78 (top 4.58%)57. WWW6 / Computer Networks: 1.78 (top 4.66%)58. ICCV: 1.78 (top 4.75%)59. IEEE Real-Time Systems Symposium: 1.78 (top 4.83%)60. AES Candidate Conference: 1.77 (top 4.91%)61. KR: 1.76 (top 4.99%)62. TISSEC: 1.76 (top 5.07%)63. ACM Conference on Electronic Commerce: 1.75 (top 5.15%)64. TOIS: 1.75 (top 5.24%)65. PEPM: 1.74 (top 5.32%)66. SIGMOD Conference: 1.74 (top 5.40%)67. Formal Methods in System Design: 1.74 (top 5.48%)68. Mobile Agents: 1.73 (top 5.56%)69. REX Workshop: 1.73 (top 5.65%)70. NMR: 1.73 (top 5.73%)71. Computing Systems: 1.72 (top 5.81%)72. LOPLAS: 1.72 (top 5.89%)73. STOC: 1.69 (top 5.97%)74. Distributed Computing: 1.69 (top 6.06%)75. KDD: 1.68 (top 6.14%)76. Symposium on Testing, Analysis, and Verification: 1.65 (top 6.22%)77. Software Development Environments (SDE): 1.64 (top 6.30%)78. SIAM J. Comput.: 1.64 (top 6.38%)79. CRYPTO: 1.63 (top 6.47%)80. Multimedia Systems: 1.62 (top 6.55%)81. ICFP: 1.62 (top 6.63%)82. Lisp and Symbolic Computation: 1.61 (top 6.71%)83. ECP: 1.61 (top 6.79%)84. CHI: 1.61 (top 6.87%)86. ACM Symposium on User Interface Software and Technology: 1.59 (top 7.04%)87. ESOP: 1.58 (top 7.12%)88. ECCV: 1.58 (top 7.20%)89. ACM Transactions on Graphics: 1.57 (top 7.28%)90. CSCW: 1.57 (top 7.37%)91. AOSE: 1.57 (top 7.45%)92. ICCL: 1.57 (top 7.53%)93. Journal of Functional Programming: 1.57 (top 7.61%)94. RTSS: 1.57 (top 7.69%)95. ECSCW: 1.56 (top 7.78%)96. TOCHI: 1.56 (top 7.86%)97. ISCA: 1.56 (top 7.94%)98. SIGMETRICS/Performance: 1.56 (top 8.02%)99. IWMM: 1.55 (top 8.10%)100. JICSLP: 1.54 (top 8.19%)101. Automatic Verification Methods for Finite State Systems: 1.54 (top 8.27%)102. WWW: 1.54 (top 8.35%)103. IEEE Transactions on Pattern Analysis and Machine Intelligence: 1.54 (top 8.43%) 104. AIPS: 1.53 (top 8.51%)105. IEEE Transactions on Visualization and Computer Graphics: 1.53 (top 8.59%) 106. VLDB: 1.52 (top 8.68%)107. Symposium on Computational Geometry: 1.51 (top 8.76%)108. FOCS: 1.51 (top 8.84%)109. A TAL: 1.51 (top 8.92%)110. SODA: 1.51 (top 9.00%)111. PPCP: 1.50 (top 9.09%)112. AAAI: 1.49 (top 9.17%)113. COLT: 1.49 (top 9.25%)114. USENIX Summer: 1.49 (top 9.33%)115. Information and Computation: 1.48 (top 9.41%)116. Java Grande: 1.47 (top 9.50%)117. ISMM: 1.47 (top 9.58%)118. ICLP/SLP: 1.47 (top 9.66%)119. SLP: 1.45 (top 9.74%)120. Structure in Complexity Theory Conference: 1.45 (top 9.82%)121. IEEE Transactions on Multimedia: 1.45 (top 9.90%)122. Rules in Database Systems: 1.44 (top 9.99%)123. ACL: 1.44 (top 10.07%)124. CONCUR: 1.44 (top 10.15%)125. SPAA: 1.44 (top 10.23%)126. J. Algorithms: 1.42 (top 10.31%)127. DOOD: 1.42 (top 10.40%)128. ESEC / SIGSOFT FSE: 1.41 (top 10.48%)130. Advances in Petri Nets: 1.41 (top 10.64%)131. ICNP: 1.40 (top 10.72%)132. SSD: 1.39 (top 10.81%)133. INFOCOM: 1.39 (top 10.89%)134. IEEE Symposium on Security and Privacy: 1.39 (top 10.97%)135. Cognitive Science: 1.38 (top 11.05%)136. TSE: 1.38 (top 11.13%)137. Storage and Retrieval for Image and Video Databases (SPIE): 1.38 (top 11.22%) 138. NACLP: 1.38 (top 11.30%)139. SIGMETRICS: 1.38 (top 11.38%)140. JACM: 1.37 (top 11.46%)141. PODC: 1.37 (top 11.54%)142. International Conference on Supercomputing: 1.36 (top 11.62%)143. Fast Software Encryption: 1.35 (top 11.71%)144. IEEE Visualization: 1.35 (top 11.79%)145. SAS: 1.35 (top 11.87%)146. TACS: 1.35 (top 11.95%)147. International Journal of Computer Vision: 1.33 (top 12.03%)148. JCSS: 1.32 (top 12.12%)149. Algorithmica: 1.31 (top 12.20%)150. TOCL: 1.30 (top 12.28%)151. Information Hiding: 1.30 (top 12.36%)152. Journal of Automated Reasoning: 1.30 (top 12.44%)153. ECCV (1): 1.29 (top 12.53%)154. PCRCW: 1.29 (top 12.61%)155. Journal of Logic and Computation: 1.29 (top 12.69%)156. KDD Workshop: 1.28 (top 12.77%)157. ML: 1.28 (top 12.85%)158. ISSTA: 1.28 (top 12.94%)159. EUROCRYPT: 1.27 (top 13.02%)160. PDIS: 1.27 (top 13.10%)161. Hypertext: 1.27 (top 13.18%)162. IWDOM: 1.27 (top 13.26%)163. PARLE (2): 1.26 (top 13.34%)164. Hybrid Systems: 1.26 (top 13.43%)165. American Journal of Computational Linguistics: 1.26 (top 13.51%)166. SPIN: 1.25 (top 13.59%)167. ICDE: 1.25 (top 13.67%)168. FMCAD: 1.25 (top 13.75%)169. SC: 1.25 (top 13.84%)170. EDBT: 1.25 (top 13.92%)171. Computational Complexity: 1.25 (top 14.00%)172. International Journal of Computatinal Geometry and Applications: 1.25 (top 14.08%)174. IJCAI (2): 1.24 (top 14.25%)175. TACAS: 1.24 (top 14.33%)176. Ubicomp: 1.24 (top 14.41%)177. MPC: 1.24 (top 14.49%)178. AWOC: 1.24 (top 14.57%)179. TLCA: 1.23 (top 14.66%)180. Emergent Neural Computational Architectures Based on Neuroscience: 1.23 (top 14.74%) 181. CADE: 1.22 (top 14.82%)182. PROCOMET: 1.22 (top 14.90%)183. ACM Multimedia: 1.22 (top 14.98%)184. IEEE Journal on Selected Areas in Communications: 1.22 (top 15.06%)185. Science of Computer Programming: 1.22 (top 15.15%)186. LCPC: 1.22 (top 15.23%)187. CT-RSA: 1.22 (top 15.31%)188. ICLP: 1.21 (top 15.39%)189. Financial Cryptography: 1.21 (top 15.47%)190. DBPL: 1.21 (top 15.56%)191. AAAI/IAAI: 1.20 (top 15.64%)192. Artificial Life: 1.20 (top 15.72%)193. Higher-Order and Symbolic Computation: 1.19 (top 15.80%)194. TKDE: 1.19 (top 15.88%)195. ACM Computing Surveys: 1.19 (top 15.97%)196. Computational Geometry: 1.18 (top 16.05%)197. Autonomous Agents and Multi-Agent Systems: 1.18 (top 16.13%)198. EWSL: 1.18 (top 16.21%)199. Learning for Natural Language Processing: 1.18 (top 16.29%)200. TAPOS: 1.17 (top 16.38%)201. TAPSOFT, Vol.1: 1.17 (top 16.46%)202. International Journal of Computational Geometry and Applications: 1.17 (top 16.54%) 203. TAPSOFT: 1.17 (top 16.62%)204. IEEE Transactions on Parallel and Distributed Systems: 1.17 (top 16.70%)205. Heterogeneous Computing Workshop: 1.16 (top 16.78%)206. Distributed and Parallel Databases: 1.16 (top 16.87%)207. DAC: 1.16 (top 16.95%)208. ICTL: 1.16 (top 17.03%)209. Performance/SIGMETRICS Tutorials: 1.16 (top 17.11%)210. IEEE Computer: 1.15 (top 17.19%)211. IEEE Real Time Technology and Applications Symposium: 1.15 (top 17.28%)212. TAPSOFT, Vol.2: 1.15 (top 17.36%)213. ACM Workshop on Role-Based Access Control: 1.15 (top 17.44%)214. WCRE: 1.14 (top 17.52%)215. Applications and Theory of Petri Nets: 1.14 (top 17.60%)216. ACM SIGOPS European Workshop: 1.14 (top 17.69%)218. Mathematical Structures in Computer Science: 1.14 (top 17.85%)219. Workshop on the Management of Replicated Data: 1.13 (top 17.93%)220. ECCV (2): 1.13 (top 18.01%)221. PPSN: 1.13 (top 18.09%)222. Middleware: 1.13 (top 18.18%)223. OODBS: 1.12 (top 18.26%)224. Electronic Colloquium on Computational Complexity (ECCC): 1.12 (top 18.34%) 225. UML: 1.12 (top 18.42%)226. Real-Time Systems: 1.12 (top 18.50%)227. FME: 1.12 (top 18.59%)228. Evolutionary Computing, AISB Workshop: 1.11 (top 18.67%)229. IEEE Conference on Computational Complexity: 1.11 (top 18.75%)230. IOPADS: 1.11 (top 18.83%)231. IJCAI: 1.10 (top 18.91%)232. ISWC: 1.10 (top 19.00%)233. SIGIR: 1.10 (top 19.08%)234. Symposium on LISP and Functional Programming: 1.10 (top 19.16%)235. PASTE: 1.10 (top 19.24%)236. HPDC: 1.10 (top 19.32%)237. Application and Theory of Petri Nets: 1.09 (top 19.41%)238. ICCAD: 1.09 (top 19.49%)239. Category Theory and Computer Science: 1.08 (top 19.57%)240. Recent Advances in Intrusion Detection: 1.08 (top 19.65%)241. JIIS: 1.08 (top 19.73%)242. TODAES: 1.08 (top 19.81%)243. Neural Computation: 1.08 (top 19.90%)244. CCL: 1.08 (top 19.98%)245. SIGPLAN Workshop: 1.08 (top 20.06%)246. DPDS: 1.07 (top 20.14%)247. ACM Multimedia (1): 1.07 (top 20.22%)248. MAAMAW: 1.07 (top 20.31%)249. Computer Graphics Forum: 1.07 (top 20.39%)250. HUG: 1.06 (top 20.47%)251. Hybrid Neural Systems: 1.06 (top 20.55%)252. SRDS: 1.06 (top 20.63%)253. TPCD: 1.06 (top 20.72%)254. ILP: 1.06 (top 20.80%)255. ARTDB: 1.06 (top 20.88%)256. NIPS: 1.06 (top 20.96%)257. Formal Aspects of Computing: 1.06 (top 21.04%)258. ECHT: 1.06 (top 21.13%)259. ICMCS: 1.06 (top 21.21%)260. Wireless Networks: 1.05 (top 21.29%)261. Advances in Data Base Theory: 1.05 (top 21.37%)262. WDAG: 1.05 (top 21.45%)263. ALP: 1.05 (top 21.53%)264. TARK: 1.05 (top 21.62%)265. PATAT: 1.05 (top 21.70%)266. ISTCS: 1.04 (top 21.78%)267. Concurrency - Practice and Experience: 1.04 (top 21.86%)268. CP: 1.04 (top 21.94%)269. Computer Vision, Graphics, and Image Processing: 1.04 (top 22.03%) 270. FTCS: 1.04 (top 22.11%)271. RTA: 1.04 (top 22.19%)272. COORDINATION: 1.03 (top 22.27%)273. CHDL: 1.03 (top 22.35%)274. Theory of Computing Systems: 1.02 (top 22.44%)275. CTRS: 1.02 (top 22.52%)276. COMPASS/ADT: 1.02 (top 22.60%)277. TOMACS: 1.02 (top 22.68%)278. IEEE Micro: 1.02 (top 22.76%)279. IEEE PACT: 1.02 (top 22.85%)280. ASIACRYPT: 1.01 (top 22.93%)281. MONET: 1.01 (top 23.01%)282. WWW7 / Computer Networks: 1.01 (top 23.09%)283. HUC: 1.01 (top 23.17%)284. Expert Database Conf.: 1.00 (top 23.25%)285. Agents: 1.00 (top 23.34%)286. CPM: 1.00 (top 23.42%)287. SIGPLAN Symposium on Compiler Construction: 1.00 (top 23.50%) 288. International Conference on Evolutionary Computation: 1.00 (top 23.58%) 289. TAGT: 1.00 (top 23.66%)290. Workshop on Parallel and Distributed Simulation: 1.00 (top 23.75%) 291. FTRTFT: 1.00 (top 23.83%)292. TPHOLs: 1.00 (top 23.91%)293. Intelligent User Interfaces: 0.99 (top 23.99%)294. Journal of Functional and Logic Programming: 0.99 (top 24.07%)295. Cluster Computing: 0.99 (top 24.16%)296. ESA: 0.99 (top 24.24%)297. PLILP: 0.99 (top 24.32%)298. COLING-ACL: 0.98 (top 24.40%)299. META: 0.97 (top 24.48%)300. IEEE MultiMedia: 0.97 (top 24.57%)301. ICALP: 0.97 (top 24.65%)302. IATA: 0.97 (top 24.73%)303. FPGA: 0.97 (top 24.81%)304. EuroCOLT: 0.97 (top 24.89%)305. New Generation Computing: 0.97 (top 24.97%)306. Automated Software Engineering: 0.97 (top 25.06%)307. GRID: 0.97 (top 25.14%)308. ISOTAS: 0.96 (top 25.22%)309. LPNMR: 0.96 (top 25.30%)310. PLILP/ALP: 0.96 (top 25.38%)311. UIST: 0.96 (top 25.47%)312. IPCO: 0.95 (top 25.55%)313. ICPP, Vol. 1: 0.95 (top 25.63%)314. PNPM: 0.95 (top 25.71%)315. HSCC: 0.95 (top 25.79%)316. ILPS: 0.95 (top 25.88%)317. RIDE-IMS: 0.95 (top 25.96%)318. Int. J. on Digital Libraries: 0.95 (top 26.04%)319. STTT: 0.94 (top 26.12%)320. MFPS: 0.94 (top 26.20%)321. Graph-Grammars and Their Application to Computer Science: 0.93 (top 26.28%) 322. Graph Drawing: 0.93 (top 26.37%)323. VRML: 0.93 (top 26.45%)324. VDM Europe: 0.93 (top 26.53%)325. AAAI/IAAI, V ol. 2: 0.93 (top 26.61%)326. Z User Workshop: 0.93 (top 26.69%)327. Constraints: 0.93 (top 26.78%)328. SCM: 0.93 (top 26.86%)329. IEEE Software: 0.92 (top 26.94%)330. World Wide Web: 0.92 (top 27.02%)331. HOA: 0.92 (top 27.10%)332. Symposium on Reliable Distributed Systems: 0.92 (top 27.19%)333. SIAM Journal on Discrete Mathematics: 0.92 (top 27.27%)334. SMILE: 0.91 (top 27.35%)335. HART: 0.91 (top 27.43%)336. ICPP, Vol. 3: 0.91 (top 27.51%)337. FASE: 0.91 (top 27.60%)338. TCS: 0.91 (top 27.68%)339. IEEE Transactions on Information Theory: 0.91 (top 27.76%)340. C++ Conference: 0.91 (top 27.84%)341. ICSE: 0.90 (top 27.92%)342. ARTS: 0.90 (top 28.00%)343. Journal of Computational Biology: 0.90 (top 28.09%)344. SIGART Bulletin: 0.90 (top 28.17%)345. TREC: 0.89 (top 28.25%)346. Implementation of Functional Languages: 0.89 (top 28.33%)347. Acta Informatica: 0.88 (top 28.41%)348. SAIG: 0.88 (top 28.50%)349. CANPC: 0.88 (top 28.58%)350. CACM: 0.87 (top 28.66%)351. PADL: 0.87 (top 28.74%)352. Networked Group Communication: 0.87 (top 28.82%)353. RECOMB: 0.87 (top 28.91%)354. ACM DL: 0.87 (top 28.99%)355. Computer Performance Evaluation: 0.87 (top 29.07%)356. Journal of Parallel and Distributed Computing: 0.86 (top 29.15%)357. PARLE (1): 0.86 (top 29.23%)358. DISC: 0.85 (top 29.32%)359. FGCS: 0.85 (top 29.40%)360. ELP: 0.85 (top 29.48%)361. IEEE Transactions on Computers: 0.85 (top 29.56%)362. JSC: 0.85 (top 29.64%)363. LOPSTR: 0.85 (top 29.72%)364. FoSSaCS: 0.85 (top 29.81%)365. World Congress on Formal Methods: 0.85 (top 29.89%)366. CHARME: 0.84 (top 29.97%)367. RIDE: 0.84 (top 30.05%)368. APPROX: 0.84 (top 30.13%)369. EWCBR: 0.84 (top 30.22%)370. CC: 0.83 (top 30.30%)371. Public Key Cryptography: 0.83 (top 30.38%)372. CA: 0.83 (top 30.46%)373. CHES: 0.83 (top 30.54%)374. ECML: 0.83 (top 30.63%)375. LCTES/OM: 0.83 (top 30.71%)376. Information Systems: 0.83 (top 30.79%)377. IJCIS: 0.83 (top 30.87%)378. Journal of Visual Languages and Computing: 0.82 (top 30.95%)379. WADS: 0.82 (top 31.04%)380. Random Structures and Algorithms: 0.81 (top 31.12%)381. ICS: 0.81 (top 31.20%)382. Data Engineering Bulletin: 0.81 (top 31.28%)383. VDM Europe (1): 0.81 (top 31.36%)384. SW AT: 0.80 (top 31.44%)385. Nordic Journal of Computing: 0.80 (top 31.53%)386. Mathematical Foundations of Programming Semantics: 0.80 (top 31.61%)387. Architectures and Compilation Techniques for Fine and Medium Grain Parallelism: 0.80 (top 31.69%)388. KBSE: 0.80 (top 31.77%)389. STACS: 0.80 (top 31.85%)390. EMSOFT: 0.80 (top 31.94%)391. IEEE Data Engineering Bulletin: 0.80 (top 32.02%)392. Annals of Mathematics and Artificial Intelligence: 0.79 (top 32.10%)393. WOSP: 0.79 (top 32.18%)394. VBC: 0.79 (top 32.26%)395. RTDB: 0.79 (top 32.35%)396. CoopIS: 0.79 (top 32.43%)397. Combinatorica: 0.79 (top 32.51%)398. International Journal of Geographical Information Systems: 0.78 (top 32.59%)399. Autonomous Robots: 0.78 (top 32.67%)400. IW-MMDBMS: 0.78 (top 32.76%)401. ESEC: 0.78 (top 32.84%)402. WADT: 0.78 (top 32.92%)403. CAAP: 0.78 (top 33.00%)404. LCTES: 0.78 (top 33.08%)405. ZUM: 0.77 (top 33.16%)406. TYPES: 0.77 (top 33.25%)407. Symposium on Reliability in Distributed Software and Database Systems: 0.77 (top 33.33%) 408. TABLEAUX: 0.77 (top 33.41%)409. International Journal of Parallel Programming: 0.77 (top 33.49%)410. COST 237 Workshop: 0.77 (top 33.57%)411. Data Types and Persistence (Appin), Informal Proceedings: 0.77 (top 33.66%)412. Evolutionary Programming: 0.77 (top 33.74%)413. Reflection: 0.76 (top 33.82%)414. SIGMOD Record: 0.76 (top 33.90%)415. Security Protocols Workshop: 0.76 (top 33.98%)416. XP1 Workshop on Database Theory: 0.76 (top 34.07%)417. EDMCC: 0.76 (top 34.15%)418. DL: 0.76 (top 34.23%)419. EDAC-ETC-EUROASIC: 0.76 (top 34.31%)420. Protocols for High-Speed Networks: 0.76 (top 34.39%)421. PPDP: 0.75 (top 34.47%)422. IFIP PACT: 0.75 (top 34.56%)423. LNCS: 0.75 (top 34.64%)424. IWQoS: 0.75 (top 34.72%)425. UK Hypertext: 0.75 (top 34.80%)426. Selected Areas in Cryptography: 0.75 (top 34.88%)427. ICATPN: 0.74 (top 34.97%)428. Workshop on Computational Geometry: 0.74 (top 35.05%)429. Integrated Network Management: 0.74 (top 35.13%)430. ICGI: 0.74 (top 35.21%)431. ICPP (2): 0.74 (top 35.29%)432. SSR: 0.74 (top 35.38%)433. ADB: 0.74 (top 35.46%)434. Object Representation in Computer Vision: 0.74 (top 35.54%)435. PSTV: 0.74 (top 35.62%)436. CAiSE: 0.74 (top 35.70%)437. On Knowledge Base Management Systems (Islamorada): 0.74 (top 35.79%)438. CIKM: 0.73 (top 35.87%)439. WSA: 0.73 (top 35.95%)440. RANDOM: 0.73 (top 36.03%)441. KRDB: 0.73 (top 36.11%)442. ISSS: 0.73 (top 36.19%)443. SIGAL International Symposium on Algorithms: 0.73 (top 36.28%)444. INFORMS Journal on Computing: 0.73 (top 36.36%)445. Computer Supported Cooperative Work: 0.73 (top 36.44%)446. CSL: 0.72 (top 36.52%)447. Computational Intelligence: 0.72 (top 36.60%)448. ICCBR: 0.72 (top 36.69%)449. ICES: 0.72 (top 36.77%)450. AI Magazine: 0.72 (top 36.85%)451. JELIA: 0.72 (top 36.93%)452. VR: 0.71 (top 37.01%)453. ICPP (1): 0.71 (top 37.10%)454. RIDE-TQP: 0.71 (top 37.18%)455. ISA: 0.71 (top 37.26%)456. Data Compression Conference: 0.71 (top 37.34%)457. CIA: 0.71 (top 37.42%)458. COSIT: 0.71 (top 37.51%)459. IJCSLP: 0.71 (top 37.59%)460. DISCO: 0.71 (top 37.67%)461. DKE: 0.71 (top 37.75%)462. IWAN: 0.71 (top 37.83%)463. Operating Systems Review: 0.70 (top 37.91%)464. IEEE Internet Computing: 0.70 (top 38.00%)465. LISP Conference: 0.70 (top 38.08%)466. C++ Workshop: 0.70 (top 38.16%)467. SPDP: 0.70 (top 38.24%)468. Fuji International Symposium on Functional and Logic Programming: 0.69 (top 38.32%) 469. LPAR: 0.69 (top 38.41%)470. ECAI: 0.69 (top 38.49%)471. Hypermedia: 0.69 (top 38.57%)472. Artificial Intelligence in Medicine: 0.69 (top 38.65%)473. AADEBUG: 0.69 (top 38.73%)474. VL: 0.69 (top 38.82%)475. FSTTCS: 0.69 (top 38.90%)476. AMAST: 0.68 (top 38.98%)477. Artificial Intelligence Review: 0.68 (top 39.06%)478. HPN: 0.68 (top 39.14%)479. POS: 0.68 (top 39.23%)480. Research Directions in High-Level Parallel Programming Languages: 0.68 (top 39.31%) 481. Performance Evaluation: 0.68 (top 39.39%)482. ASE: 0.68 (top 39.47%)483. LFCS: 0.67 (top 39.55%)484. ISCOPE: 0.67 (top 39.63%)485. Workshop on Software and Performance: 0.67 (top 39.72%)486. European Workshop on Applications and Theory in Petri Nets: 0.67 (top 39.80%) 487. ICPP: 0.67 (top 39.88%)488. INFOVIS: 0.67 (top 39.96%)489. Description Logics: 0.67 (top 40.04%)490. JCDKB: 0.67 (top 40.13%)491. Euro-Par, Vol. I: 0.67 (top 40.21%)492. RoboCup: 0.67 (top 40.29%)493. Symposium on Solid Modeling and Applications: 0.66 (top 40.37%)494. TPLP: 0.66 (top 40.45%)495. CVRMed: 0.66 (top 40.54%)496. POS/PJW: 0.66 (top 40.62%)497. FORTE: 0.66 (top 40.70%)498. IPMI: 0.66 (top 40.78%)499. ADL: 0.66 (top 40.86%)。

Digital self-service is the new normal. Customers expect compelling, convenient digital experiences any time, on any device—and they are demanding the same from the financial services industry,government agencies and other regulated businesses. To meet the demand, these organizations are now making complex, high-value transactions such as mortgage applications, benefits enrollments and more easily accessible onto digital channels—from web or mobile.Creating streamlined, personalized digital enrollment, onboarding and communications significantly accelerates revenue, cost efficiency and client satisfaction. With 70% of FSI executives believing mobile will be a major source of new accounts in the next three years, offering mobile-friendly enrollment to all clients on all devices is now an imperative.1 Digitizing transactions and communications also eliminates paper-based costs. Printing, distribution and storage of paper disappears, while low value manual tasks like data rekeying and document routing can also be streamlined. As the quality of digital experiences increases, so does adoption of more efficient digital channels over costly alternatives such as in-person, mail, or call center based communication. A Deloitte report found that the average in-person transaction costs a government agency almost 17 dollars, while a digital transaction for the same process cost merely 40 cents.2 With digital being over 40 times more cost efficient than in-person processing, it's no surprise that organizations across regulated industries are prioritizing digital self-service as part of their transformation initiatives.1. Digital Trends in Financial Services Industry, eConsultancy & Adobe, 20172. Deloitte Access Economics, Digital Government Transformation, Commissioned by Adobe, 2015Key Capabilities• Author—empower business users to quickly design, approve, and publish centrally-managed forms and communications.• Discover—help users quickly find relevant forms and documents through search, rules, filters, and even geolocation.• Enroll—streamline enrollment with mobile-optimized adaptive forms to reduce abandonment.• Process—digitize and automate onboarding processes through visually-designed workflows.• Communicate—design, generate, and delivery multi-channel,personalized communications to increase engagement and retention.• Secure—protect your valuable documents through customizable role and access policies and encryption, even outside of your organization’s firewall.• Improve—optimize your enrollment experiences with granular analytics and targeted content to personalize the experience to each customer.Adobe Experience Manager FormsStreamline enrollment, onboarding and multi-channel communications with digital forms and documents.Adobe Experience ManagerA state development bank reduced processing times by over 50%.Learn howA global bank increased conversion rates on some applications from 33 to 80%.Learn howAdobe Experience Manager FormsAdobe Experience Manager Forms helps organizations deliver secure, streamlined application enrollment across any device, process submissions efficiently with automated workflows, and enable personalized onboarding and communications for standout customer experience that increases conversion, retention, and client satisfaction.Businesses and government agencies use Experience Manager Forms to:• Accelerate digital enrollment conversion by pre-filling form fields from back-end systems, enabling seamless multidevice interactions and measuring abandonment rates at form field levels.• Lower the cost of processing applications with end-to-end digital workflows and digital signatures.• Deliver multichannel interactive communications with batch and on-demand statements and welcome kits for web, print and PDF channels.• Increase business agility and scale by empowering business users to author, publish, and manage centralized form and document collections through an intuitive, drag & drop user interface.• Increase field agent productivity with offline mobile data capture and back-end data integrations.• Improve regulatory compliance by ensuring secure transmission of personally identifiable information (PII)in forms, generating documents of record for audit trails, encrypting sensitive documents, and aligning with accessibility standards.Key features of Adobe Experience Manager FormsAuthorReach more clients faster by empowering business users to author, publish, and manage responsive and consistent forms and documents across multiple channels without coding.• Use a central repository and business-friendly UI for creating, editing, and managing forms and documents.• Author a form once and render it across multiple screens, channels, and formats, even print.• Drag and drop the standard components you need, such as text fields, pull-down menus, buttons, charts, electronic signature capture, and more.• Leverage out-of-the-box style themes and form and document templates, or create your own.• Create reusable form or document fragments, such as address block fields or standard paragraphs of text, and share and update across many forms or documents.• Add dynamic behavior or web services integrations to forms without coding using a visual rule editor.• Debug forms quickly with built-in development tooling.• Preview how forms and documents will look on different devices and screens before publishing.• Send form and document templates for content and branding review by stakeholders using customizable, automated workflows.• Localize forms and documents by taking advantage of workflows that connect to machine or human translation services.Get offers to market quickly by leveraging themes, templates, drag-and-drop content authoring, and more. DiscoverHelp organizations connect clients with the form they need, streamlining the digital enrollment journey.• Drag and drop a forms portal into your existing websites or ones built with Experience Manager Sites.• Allow clients to search for forms in the portal using keywords, tags, or other properties such as date last modified.• Embed links to abandoned forms via an email campaign delivered using Adobe Campaign to re-targetlost customers.Facilitate form discovery for your customers to accelerate enrollment and decrease customer support costs.EnrollMake easy for customers to complete forms on any device quickly and error-free.• Organize forms into sections that update based on user input to simplify filling on all devices.• Support for accessibility standards including Section 508 and WCAG.• Pre-fill form fields through easy integration with CRM systems, social logins, or use web services to populate fields based on user input.• Reduce keystrokes by taking advantage of device features, such as camera, barcode scanning, or speech-to-text.• Capture secure, legal, and compliant e-signatures using native integration with Adobe Sign.• Validate form fields such as phone and address and offer context-sensitive Help to reduce the possibilityof errors.• Allow clients to save in-progress forms and return later to complete them, even on another device.• Verify submissions with CAPTCHA support.Simplify form filling by using device cameras to read barcodes and fill in information via web services. ProcessProcess submissions quickly by connecting form data with existing back office systems, business rules, workflows and people.• Build digital workflows to process submitted form applications easily with a drag-and-drop interface.• Perform actions such as view assigned tasks, track progress, review, approve, reject, and more.• Use a customizable mobile app that allows mobile workers to securely collect and record data across multiple forms on tablets or smartphones, even when offline.• Speed integration with your back-end and third-party applications with a data integration tool, create a form data model, or leverage out-of-the-box connectors to popular RDBMS and CRM systems.• Support complex e-signature requirements with Adobe Sign, including multiple signers, sequential and parallel signing workflows, anonymous user signing, and verifying signer identity.• Leverage document generation workflows to automatically generate, deliver, and archive branded PDF documents of record.The Experience Manager Forms App allows field workers to securely captureinformation and submit data via forms while on the go—even offline. CommunicateImprove client retention and satisfaction with personalized, interactive and engaging communications and documents. • Create and approve communications letters quickly with a business-user friendly agent interface that allows agents to author letters using templates, preapproved content blocks, business rules, and more—for web, print and PDF channels.• Generate engaging Interactive Communications in responsive documents, including dynamic charts and personalized marketing offers to drive effective upsell and cross-sell.• Connect back-end data sources to automatically personalize communications at scale.• Preview responsive documents for different devices before publishing.• Support on-demand or automated batch document delivery through multiple channels, including web portals, mobile apps, PDF, email, and paper.• Optionally validate content integrity and signer’s identity with digital signatures for PDF documents.• Automatically generate, delivery, and archive branded customer communicationsInteractive Communications features personalize engagement and improve retention.SecureProtect sensitive information contained in PDF and Microsoft Office documents based on business policy, even outside of your organization's firewall.3• Use strong encryption to protect information contained in documents.• Tailor document access and usage rights to the sensitivity of the information, and track its use.• Protect bulk or system-generated documents, such as statements.• Revoke usage rights at any time—even if the document has been distributed outside the organization.• Work in conjunction with strong user authentication systems, including single sign-on (SSO), security assertion markup language (SAML), and public key infrastructure (PKI).• Store protected documents through integration with Experience Manager Assets or ECM systems.• Visually track document usage and detect anomalies with the Data Workbench capability in Adobe Analytics. OptimizeWork with other Adobe Experience Cloud solutions to continually improve customer experiences, maximizing conversion rates and satisfaction.• Understand how users interact with forms and documents using out-of-the-box reports in Adobe Analytics.• Drill down into report details to pinpoint form abandonment at the form field level, generating actionable insights to improve conversion.• Use A/B testing in Adobe Target to test new form or document versions, monitor the test, and automatically publish better performing versions.• Engage customers and increase cross-sell opportunities by inserting personalized and dynamic content into responsive forms and documents with Experience Targeting, powered by Adobe Target.• Accelerate going paperless with Automated Forms Conversion powered by machine learning and Adobe Sensei. Convert legacy PDF forms to responsive forms, extract reusable fragments, and continuously improveform collections.Gain insight into form or document usage with Adobe Analytics reports.3. Requires purchase of Document Security add-onBeyond product innovation: Transform the customer journeyAdobe is committed to developing industry-leading solutions. We also offer essential resources to help you transform your entire organization to deliver better digital experiences. Here are some key resources to help you lead and succeed.Adobe, the Adobe logo, and Acrobat are either registered trademarks or trademarks of Adobe Systems Incorporated in the United States and/or other countries. All other trademarks are the property of their respective owners.© 2018 Adobe Systems Incorporated. All rights reserved.For more information/go/aemAbout Adobe Experience ManagerAdobe Experience Manager allows marketers and developers to create, manage, and deliver customer-facing digital experiences across everychannel—web, mobile, social, video, in-store, and IoT. The solution spans digital experience management including web and mobile, digital asset management, social communities, and forms and documents. Adobe Experience Manager integrates with other Adobe solutions, allowing businesses to use data insights to deliver targeted content to identified segments and transform content into engaging, personalized experiences—optimizing customer engagement and lead generation and accelerating revenue.。

Oracle9i安装过程RHEL4U4RedHat as4 u3 + Oracle 9.2.0.4 刘兴广 2008-09-06一：环境配置1，创建oracle安装用户组及用户帐号；#groupadd dba#groupadd oinstall#useradd oracle -g oinstall -G dba#passwd oracle2，建立oracle安装文件夹；# mkdir -p /app/oracle/product/9.2.0.4# mkdir /app/cwdata# chmod 777 /app/cwdata# chown oracle.dba /app/cwdata# chown oracle.dba /install (放置安装文件的地方)# chmod 777 /install# chown -R oracle.dba /app/oracle/product/9.2.0.4# chmod -R 777 /app/oracle/product/9.2.0.42，配置环境变量；以root用户登录，设置root用户的环境打开.bash_profile文件，追加如下内容加入：## for oracle addexport ORACLE_BASE=/app/oracleexport ORACLE_HOME=/app/oracle/product/9.2.0.4export PA TH=$PA TH ORACLE_HOME/bin ORACLE_HOME/Apache/Apache/binexport ORACLE_OWNER=oracleexport ORACLE_SID=oradb3、设定oracle安装用户环境参数，以oracle用户登录，修改环境配置文件：vi /home/oracle/.bash_profile修改为：# .bash_profile# Get the aliases and functionsif [ -f ~/.bashrc ]; then. ~/.bashrcfi# User specific environment and startup programsPA TH=$PA TH HOME/binunset USERNAME## for oracle addexport ORACLE_BASE=/app/oracleexport ORACLE_HOME=$ORACLE_BASE/product/9.2.0.4export ORACLE_SID=oradbexport ORACLE_TERM=xtermLD_LIBRARY_PA TH=$ORACLE_HOME/lib:/lib:/usr/libLD_LIBRARY_PA TH=$LD_LIBRARY_PA TH:/usr/local/libexport LD_LIBRARY_PA THexport PA TH=$PA TH ORACLE_HOME/bin## end4，设置系统参数；#su – root切换到root用户a) 修改#vi /etc/sysctl.conf, 以下是配置文件的内容（其中红色字体为添加的内容）：# Kernel sysctl configuration file for Red Hat Linux## For binary values, 0 is disabled, 1 is enabled. See sysctl(and# sysctl.conf(5) for more details.# Controls IP packet forwardingnet.ipv4.ip_forward = 0# Controls source route verificationnet.ipv4.conf.default.rp_filter = 1# Controls the System Request debugging functionality of the kernelkernel.sysrq = 0# Controls whether core dumps will append the PID to the core filename.# Useful for debugging multi-threaded applications.kernel.core_uses_pid = 1kernel.shmmax = 536870912kernel.shmmni = 4096kernel.shmall = 2097152kernel.sem = 250 32000 100 128fs.file-max = 65536net.ipv4.ip_local_port_range = 1024 65000修改后运行#sysctl –p命令使得内核改变立即生效；译者注：一般情况下可以设置最大共享内存为物理内存的一半，如果物理内存是2G，则可以设置最大共享内存为1073741824，如上；如物理内存是1G，则可以设置最大共享内存为512 * 1024 * 1024 = 536870912；以此类推。

导师布置的第一个任务：jade安装，跑实例。

（一周时间）jade（Java Agent DEvelopment framework）是基于java语言的开发agent的工作框架。

也可以说是用java开发的一个开发agent的软件；或者说是用于开发符合FIPA规范的多agent系统的平台。

第一步安装，到官方网站下载最新版本的jade（目前是3.5），在安装jade前要先安装好jsk。

1、文件结构：解压后jadeAll3.5文件夹下有四个文件夹分别是JADE －bin－3.5（存放jade的核心部件）、JADE‐ doc－3.5（存放jade的说明文档）、JADE‐examples－3.5（存放jade的实例源代码）、JADE‐src－3.5（存放资源文 件，这个我还没搞清楚）2、classpath设置：要包含在JADE‐BIN‐3.5\jade\lib\目录下有http、iiop、jade、jadeTools和commons－codec－1.3这五个文件。

如我自己设置的classpath为例.D:\Program Files\Java\jadeAll3.5\JADE‐bin‐3.5\jade\lib\http.jar;D:\ProgramFiles\Java\jadeAll3.5\JADE‐bin‐3.5\jade\lib\iiop.jar;D:\ProgramFiles\Java\jadeAll3.5\JADE‐bin‐3.5\jade\lib\jade.jar;D:\ProgramFiles\Java\jadeAll3.5\JADE‐bin‐3.5\jade\lib\jadeTools.jar;D:\ProgramFiles\Java\jadeAll3.5\JADE‐bin‐3.5\jade\lib\comons‐codec\commons‐codec ‐1.3.jar注：classpath就是让系统自己能找到jade的程序。

Mac-A-Mal:An Automated Framework for Mac Malware HuntingPham Duy Phuc*************************SfyLabsFabio Massacci*********************** University of TrentoAbstractAs Mac systems grow in popularity,so does macOS mal-ware whilst macOS malware analysis is still lagging be-hind,even when researchers deal with malicious behav-iors in the user space.To amend this shortcoming,we have come up with macOS analyzer for malware Mac-A-Mal:A system for behavioral monitoring of components at kernel level which allows analysts to automatically in-vestigate malware on macOS,broadly extending what is available today with Cuckoo sandbox.By leveraging on kernel-level system calls hooking,the framework is able to detect and mitigate malware anti-analysis tech-niques.In particular,it combines static and dynamic analysis to extract useful information and suspicious be-haviors from malware binaries,their monitored behaviors such as network traffic,malware evasion techniques,per-sistence methods,file operations etc.,without being de-tected by common Mac malware evasion techniques.We have used the framework to evaluate thousands macOS samples to estimate how widespread Mac malware vari-ants and families are today(thanks to VirusTotal).Mac malware in2017demonstrates a drastic improvement by using evasion techniques.Overall,we used our systems to classify the dataset and found that85%of collected samples are adware,49%of classified variants belong to backdoor/trojan.By hunting Mac samples on Virus-Total,we found a so-far-undiscovered organized adware campaign which leverages several Apple legitimate de-veloper certificates,a few of other undetected keyloggers, and trojan samples participating in APT32OceanLotus targeting Chinese and Vietnamese organizations,as well as hundreds of malware samples which have otherwise low detection rates.1IntroductionContrary to popular belief,Mac is not immune from mal-ware.According to a cyber-security report from Bit9and Carbon Black Threat Research team,Mac devices have seen more malware attacks in2015than the pastfive years combined.Mac malware grew744%with around 460,000instances detected,says McAfee report in2016. The trend in Mac malware was not slowing down in2017, there were nearly300,000new instances of macOS mal-ware in thefirst3quarters of2017,lifting the year-total to over700,000.[6]The malware problem has become bigger and stronger over years,while the majority of malware targets at Mi-crosoft Windows operating system,other operating sys-tems have become relevant targets as well.By under-standing how they behave,what these threats are,and engaging in ways to detect them,we can contribute to more secure and stable solutions for Mac computing ex-perience.However,research about macOS malware and relevant solutions to automated Mac malware analysis are limited.While many types of state-of-the-art malware on Windows platform took decades to develop from the first known malicious software to happen in the wild,now they start emerging on Apple computers in a shorter time. Static malware analyses are mostly used to analyze macOS malware. E.Walkup[10]was able to extract Mach-O structure,import libraries(DyLib)and functions feature from a data set consisting of420malware samples and1000goodware over20machine learning models. Furthermore,S.Hsieh et al.[4]presented a study of clas-sifying Mac OS X malware in2016with a set of features extracted from Mach-O metadata using tools such as nm, otool,or strings on VirusTotal sample collection of 2015-2016.They also included derivative numerical fea-tures created from meta information,which are introduced in learning-based malware classification,e.g.function call distribution,structure complexity,etc.D.Dorsey[3] presented a weighted distance metric solution to generate and compare assembly mnemonics signature of Mach-O binaries using mpesm.The study successfully discovered 7over18,000samples to be UPX compressed,and3 groups signature matched at least85%of malicious sam-ples.However,no significant signature could identify the difference between malicious samples and17,000known benign samples.Regarding dynamic Mac malware studies,V. Mieghem[8,9]presented a novel generic behavioral detection method based on system calls names and prevention mechanism for malware on OS X.Sequences of system call traces are analyzed,from which certain malicious system call patterns,interactions with shells and auto-run services appear to be an accurate indication of malware on a system.Three types of user profiles are established to evaluate the detection patterns,resulting in a100%detection rate and a0%to20%false positive rate, depending on the type of user profile.M.Lindorfer et al.[5]have built an OS X honeypot based high-interaction and used it to evaluate more than6thousands blacklisted URLs to estimate how widespread malware for Mac OS X is in2013.Onlyfive websites were found to drop binaries through drive-by downloads,but none of them targeted OS X.Furthermore,they developed a dynamic analysis environment with DTrace and analyzed148 malicious samples.The result has shown that while some OS X malware families are sophisticated,several fail to perform simple but critical jobs like persistence. A. Case et al.[1,2]presented malware detection techniques with a specific focus on kernel-mode components for OS X,particularly for rootkit that targets at kernel data and user land malware written in Objective-C for Mac.They utilize V olatility to detect rookit in kernel memory.For Objective-C malicious code,they analyzed important artifacts in the memory and produced output that could easily be used by analysts to isolate and investigate more deeply these behaviors even when the Objective-C runtime maintained state out of the dynamic loader and the code section of executables.We evaluated some of popular dynamic macOS analyz-ers and we found that Cuckoo sandbox is an open source project providing most robust solution for malware analy-sis framework.Although it only supports Windows,Linux but partially Android and macOS.It adopts DTrace which is basically a binary instrumentation platform rather than a malware analysis.Apple does not allow DTrace to mon-itor official Apple binaries and requires traced software to run under root permission,its tracing techniques can be easily defeated by trivial anti-debugging tricks.Besides, Macsandbox1is a modified version of Cuckoo sandbox on Mac,using Dtrace and library injection for process tracing.However,running on user space makes it vul-nerable to trivial anti-debug and anti-hook techniques. For analysis on kernel-space region,Fireeye Monitor2 is a closed source software for manual analysis,which 1https:///sandialabs/mac-sandbox2https:///blog/threat-research/ 2017/03/introducing_monitor.html performs process execution,file and network logging.In this white paper,we study an automated solution for Mac malware analysis framework.We show that it is possible to automatically perform behavioral monitor of process execution,file activities,network traffic,with regards of virtualized environment,malware evasion de-tection and mitigation.2Mac-A-Mal:Automated macOS mal-ware analyzerMac-A-Mal is a combination of both static and dynamic ing static analysis to understand in depth sus-picious areas of code,and dynamic analysis to unpack packed binary and monitor malware behaviors.It takes actual behavioral data of malware samples executions inside virtual sandboxes simultaneously,in which it can process multiple samples at once.The sandbox is armored with network sniffer,system calls and behavior logging, as well as anti-evasion from kernel-mode to send back report to analysis machine.Researchers thereafter can review reports to spot any suspect activities such as sus-picious network activity,invoked anti debug techniques, persistent activities etc.Researchers can also perform common malware detection techniques such as Y ARA rules and behavioral signature to automatically detect a whole sample family.The overview design of Mac-A-Mal framework is illustrated in Fig.1.The analysis machine gathers samples and feed them to the monitor machine(s) running on macOS simultaneously.Data sharing features and default Apple protection mechanisms such as XPro-tect,Gatekeeper,etc.are disabled to capture malware behavior accurately.The analyzer processes2main tasks:•Static analysis:Parsing Macho executable and dis-play:Symbol table,segment,sections,load com-mands,dylib,entropy etc.Analyzing other common Macfile types such as DMG,PKG etc.as well as their certificate details.•Results collection:Output to web front-end as well as JSON format which can be easily applied with behavioral signature,machine learning or Y ARA rules.The monitor processes following tasks:•Dynamic analysis:Processing monitor behavior un-der kernel-space with Anti anti-vm,anti-debug etc.logging and mitigation.Additionally,common mal-ware analysis techniques are implemented such as network monitor,droppedfiles collection,etc.We implement an analysis agent on analysis machine un-der kernel space,which basically performs:SyscallFigure1:Mac-A-Mal framework designtable discovery technique to unslide KASLR to ob-tain kernel base image address,then walks through the kernel base image to locate load command seg-ment,andfinally performs system table lookups. These proposed techniques are compatible to latest version of macOS High Sierra.Once the sysent ta-ble is located,we patch its entries to our defined callback functions.In particular,when a malware invokes system calls,it queries the address of the call in the XNU kernel syscall table,which now is patched to our callback function address.The callback functions will perform execution logging, malware evasion detection and mitigation,as well as post-processing for the calls(forward or drop).•Anti evasion:We develop kernel system calls log-ging in which existing solutions failed,e.g.by hook-ing into posix spawn(),execve()and copying double pointers between user space and kernel space. Fig.2shows how we detect and mitigate evasion techniques by hooking into ptrace(),csrctl(), ioctl()and sysctl().•Execute samples via open default handler:By hook-ing in kernel space,we can execute samples us-ing default Mac handler-/usr/bin/open as wellas XPCProxy,in which Dtrace and MacSandbox failed to trace.It means any types offiles can be opened with capability of process forks tracing(e.g. .app;.dmg;.doc;.zip etc.)without using other app-opener.•No human interaction:During the analysis,the sand-box implements libraries for taking screenshots of the analysis screen.These screen shots are helpful for analysts to review automated analysis and recog-nize some cases that needs human interaction,for instance mouse click to confirm installation or pass-word typing for Authorization solver.We use Quartz library to detect the login window then answer pass-word using KeyboardEvents.Screenshot images are analyzed to seek for confirmation buttons.If detected,it performs mouse clicks by using Quartzs MouseEvents.If packages are stored within an Ap-ple disk image,hdiutil will be used to attach the disk,and execute all application inside its root folder.•Hardening the virtual analysis machine:We modify virtual machine footprints in machine configuration as well as manipulate its hardware information in kernel level.Besides,default Apple security featuresFigure2:Mac-a-Mal anti environment evasion module using kernel hooking and memory patching.(1)A system call is invoked in the user space,(2)Hook the system call,(3)Return the control to the original system call,(4)Returns value to the user space and manipulate system call arguments if anti-evasion attempts are detected,(5)The return value is logged for further analysis using Logger component,(6)Return values from the kernel space to the analyzer in userspace,(7)the exit()system call is not hooked and forwarded back to the user space.such as XProtect,GateKeeper,etc.are deactivated in order to execute any arbitrary malicious samples. 3Case studies of undetected Mac malware discoveryThese case studies dive deeper into the illustration drawn in Section2by explaining detailed analysis on some major malware campaigns which are found by Mac-A-Mal.•OSX/Mughthesec[7]The adware campaign wasfirst founded in the wild in August2017.It pretended to install legitimate Adobe flash and silently install Potential Unwanted Ap-plications(PUA)such as Booking,Advanced Mac Cleaner,SafeFinder Safari extension,AdBlock,etc.Reports show that the campaign dropped some ma-licious binaries related to AMCleaner adware cam-paign.It was likely an affiliation advertising cam-paign,in which adware authors spent some moneyfor buying legitimate Apple developer certificates. By using a combination of behavioral rules(antivm), static rules(suspicious legitimate Apple developer certificate),and suspect network activities based on results from Mac-A-Mal,we discovered71signed archives.Attackers had used at least10Apple legiti-mate developer certificates,and only2of them were revoked at the time of discovery.Moreover,some of dropped MachO executables were not signed,which means it could transform from adware to backdoor silently.•APT32-OceanLotusAPT32was an APT campaign targeted Chinese and Vietnamese infrastructure.It wasfirst discovered by Qihoo360,and followed up by AlienVault,Fire-Eye and Palo Alto Networks.By study thefirst generation of Mac OceanLotus samples through our framework,we generated some similar behavioral signatures amongst the family.In March2017,wefound second generation of Mac APT32which has 0detection rate over more thanfifty Anti-virus ven-dors by hunting those behaviors on VirusTotal Intelli-gence service.Ourfinding shows that the campaign was improved from thefirst variant by eliminating all Bash command executions.Unlike common Mac malware,this variant tried to execute under user privilege instead of higher privilege that may easily convey malicious behaviors.4DiscussionWe have processed more than2000malicious samples and85%of the collected samples are adware,which are dominated by OSX/Pirrit and OSX/MacKeeper.That statistic result is well confirmed according to reports from other Anti-virus vendors.Once installed,macOS adware usually persistent deep inside the victim system and starts hjacking browser.Their purpose is to make advertising revenue for attackers by installing PUA or redirecting victims to unwanted websites.They are widespread be-cause technically they are not viruses but can potentially perform malicious activities.After analyzing the samples set,results retrieved from Mac-A-Mal were later post-processed through a classi-fier based on numerous rules:code signing authority, persistence indicators,processes creation and shell exe-cution,backdoor port listening,network activities,Tor network indicator,anti-analysis techniques,file creation and modification,browsers change and super-user per-mission authentication,etc.We also compared them to VirusTotal detection results and other Anti-virus labeled variant name to verify the accuracy.We observed a total of86different Mac malware families until2017,and49% of them belongs to backdoor/trojan variant.We study the evolution of macOS malware by using heat map analysis of most Mac malware variants in Ap-pendix Fig.3and Fig. 4.We evaluate that macOS malware in2017demonstrates a drastic improvement by using anti-analysis techniques.The behaviors can be grouped into following categories:(i)persistence,file write,browser modification;(ii)root request,bash exe-cution;(iii)anti analysis;(iv)network,Tor,open port.5 variants are found applying various evasion techniques including ptrace(),sysctl(),sleep().Some vari-ants tried to discover running security tools on the system or terminate analysis software(e.g.DTrace,lldb,etc.). Besides,some samples never made any persistence at-tempt such as XAgent which is an upgraded version of Komplex in2016also known for participating in APT28 targeting individuals in the aerospace industry running macOS.It can be seen that Mac malware in2017car-ried out more frequent browser modification activities.It means that macOS malware is becoming more grey,and it has been boosted by numerous adware to redirect victims to fraud web traffic or unwanted advertising.In addition, a large number of increased super-user request behaviors shows that Mac malware is trying to increment phising user passwords in order to grant root access of the system and perform more advanced malicious activities.5Future WorkWe would like to later apply more robust and advanced techniques for better features extraction from the analysis, and machine learning for a larger scale of Mac samples. References[1]C ASE,A.,AND R ICHARD,G.G.Advancing mac os x rootkitdetection.Digital Investigation14(2015),25–33.[2]C ASE,A.,AND R ICHARD,G.G.Detecting objective-c malwarethrough memory forensics.Digital Investigation18(2016),3–10.[3]DORSEY,D.Analyzing entrypoint instruction differencesin mach-ofiles with mpesm.https://www.carbonblack.com/2016/03/01/analyzing-entrypoint-instruction-differences-in-mach-o-files-with-mpesm/,2006.Accessed:30-July-2017.[4]H SIEH,S.,W U,P.,AND L IU,H.Automatic classifying of macos x samples.Virus Bulletin,2016.Accessed:20-July-2017. [5]L INDORFER,M.,M ILLER,B.,N EUGSCHWANDTNER,M.,ANDP LATZER,C.Take a bite-finding the worm in the apple.In Infor-mation,Communications and Signal Processing(ICICS)20139th International Conference on(2013),IEEE,pp.1–5.[6]M INIHANE,N.,M ORENO,F.,P ETERSON,E.,S AMANI,R.,S CHMUGAR,C.,S OMMER,D.,AND S UN,B.Mcafee labs threats report,December2017.[7]P HUC,P.D.What is safefinder/operatormac campaign?,07/2017(accessed30-August-2017).[8]V AN M IEGHEM,V.Detecting malicious behaviour using systemcalls.TU Delft Repositories,2016.[9]V AN M IEGHEM,V.Behavioural detection and prevention ofmalware on os x.Virus Bulletin,2016(accessed20-July-2017).[10]W ALKUP,E.Mac malware detection via staticfile structureanalysis.University of Stanford,CS229:Machine Learning Final Projects,2014.A Publication of this WorkSource code of the framework is all available for down-load on Github:https:///phdphuc/mac-a-mal.The academic paper of this work has been composed in a paper”MAC-A-MAL:macOS Malware Analysis Framework Resistant to anti evasion techniques”which was submitted to the International Symposium on Engi-neering Secure Software and Systems ESSoS2018.B AppendixFakeFileOpener and MacKeeper are classified as Adware,Keranger is ransomware and the rest of variants is trojan/backdoor.Figure3:Behavior heatmap of macOS malware in2016ChromePatch and Mughthesec are classified as Adware,Macransom is ransomware and the rest of variants is trojan/backdoor.Figure4:Behavior heatmap of macOS malware in2017。

东南大学计算机网络和信息集成教育部重点实验室2012年度考核报告2011年9月，实验室接受教育部组织的现场评估，取得良好的评估成绩。

按照教育部相关实验室管理规定，完成评估意味着上一届实验室领导班子管理工作的结束，因此2012年度的实验室工作实际上是上一年度工作的自然延续，取得的成果主要体现在以下几个方面。

1、实验室科研工作围绕“计算机网络和信息集成”的主线，在体现“理论—技术—应用”关联性的基础上，致力开展计算机网络体系结构和协议、网络安全和管理、分布式计算、信息集成与现代服务等4个方向的理论、方法与关键技术的研究与开发工作。

（1）、项目承担情况在研69项，新立项25项（见表1）。

表1 实验室2012年度在研科研项目一览表（2）、论文情况2012年实验室完成论文超过200篇，3大检索约200篇（见表2）。

表2 2012年度实验室部分收录论文（3）、专利及著作权情况（见表3）2、人才培养情况研究生培养工作：继续承担研究生教学任务和博士/硕士生培养工作；本科教学工作：承担本科教学任务；指导毕业设计工作；优秀生培养。

其他类指导工作：招收工程硕士，接纳进修教师和接受博士后入站。

3、2012年度经费使用说明2012年度，学校下拨实验室运行经费32万元，使用情况见表4。

4、下一步工作计划2013年度的首要工作是实验室领导班子和学术委员会换届；对应新一届领导班子和学术委员会，建议首要考虑加强研究队伍的建设，加大中青年杰出人才的培养和引进力度；同时，加强应用基础研究，扎实开展创新性和高水平的研究工作。

东南大学计算机网络和信息集成教育部重点实验室2013年1月10日。

Modeling and Simulation of Ecosystem Based on Multi-AgentSystemZhen Li, Guojian Cheng, Xinjian QiangSchool of Computer ScienceXi’an Shiyou UniversityXi’an, Shaanxi ProvinceChinazhli@, gjcheng@, qiangxj@Abstract: - Based on the complex adaptive system theory and the multi-agent modeling and simulation method,a reactive agent model of ecosystem is built in this paper. The model tries to naturally show the general characteristics of grassland ecosystem while retained its eco-adaptive complexity and diversity. In experiment, the model is implemented on the Net Logo which is a multi-agent simulation platform. By adjusting environmental factors and various parameters, a variety of real phenomenon of a grassland ecosystem is emerged. Finally, the simulation results are analyzed and compared them with the actual ecosystem showing the correctness and expansibility of this model.Key-Words: - Multi-Agent System, Net Logo, Ecosystem, Complex Adaptive System, Modeling, Simulation1 IntroductionComplex Adaptive System, put forward firstly by the American computer scientists Holland in 1994. The basic idea is: members of CAS can constantly interact with system environment and other subjects and continuously "learning" and "experience" in this process, according to these experiences to change their structure and behavior, thus emphases the new structure , phenomenon and more complex behavior in the overall level . Complex Adaptive System's complexity stems from its main body of adaptability, namely “establishment of complex adaptive”[1]. Actually, the eco-system is a large and complex system. The Eco-system consists of a large number of populations which have nonlinear interactions between each other, these populations constitute a hierarchical system and the low-level’s disorder on the local behavior of various groups may have shown a high level of order behavior patterns; In order to survive, the various groups of ecosystems must adapt to the changing environment; All species in the ecosystem have their own information, and problem solving skills; There are also no direct order to teach each specifics ’s behavior of the various global control institutions in ecological system ; The behavior of each species is asynchronous concurrent in ecosystem. The mentioned systems nonlinear, hierarchical,self-organizing, adaptive, decentralized control, asynchronous concurrency and other characteristics, is the characteristics of complex adaptive systems. Therefore, the ecosystem is a typical time of evolving complex adaptive systems. For such complex systems, using mathematical2 Multi-Agent system’s architecture and features2.1 Agent’s architecture and featuresThe so-called Agent is a physical or abstract entity, it can act on their own and the environment, and to respond to the environment. Generally speaking, Agent has the knowledge, goals and capabilities. Knowledge refers to the Agent on its surroundings or it required description of the problem’s solution;goals refer to all acts taken by the Agent aregoal-oriented; capacity refers to the Agent with the reasoning, decision-making, planning and control capabilities.2.2 Multi-Agent SystemMulti-Agent System (MAS) is a collection which composed by multiple calculable Agent .The MAS modeling simulation is a modeling simulation method based on the object-oriented and bottom uptechnology. It uses various Agent attributes and behaviors in MAS, the interactions between the individual of simulation component system and individual, through individual attributes and behaviors with integral attributes and behaviors to feedback and correction and to study the development and evolutionary of the system [2-5].MAS have general characteristics as follows [1]:1. Each Agent has only the partial or incomplete information or with only part of capabilities to solving problem, so each Agent’s vision is limited.2. MAS without a global control system.3. Data is scattered processing and storage.4. Can realize asynchronous calculation.The above features of MAS just correspond to the generated autonomy, distribution, and parallelism features of the ecosystem. Therefore, the introduction of MAS can be used for simulation, optimization, simulation and control ecosystem. It has also become an important method and means to research the ecological system.MAS modeling’s major concepts: through the changes way of designated Agent internal state, to observe a large number of mutual Agent interaction’s results, and trying to find out emerged phenomenon in the complex environment; the global control are not present in the process of modeling and simulation, For a specific certain Agent is set simple attributes and behaviors, and then through other Agent or environmental interaction to evolve the system.3 The Multi Agent Model of Ecological system3.1 The Multi Agent Model of Ecological systemThis model mainly consists of inorganic environment Agent, producers, consumers Agent. The model constituted a local environment of ecological system. The local environment is two-dimensional, classified as composed grid of Agent producers, each producer Agent have a small piece of rectangular area, it is can't move. Agent Consumer can move freely andproduce the relevant behavior in the area constitute by Agent producers. Inorganic environment agent as a global Agent, it watched producers Agent and consumer Agent constitute local environment, can obtain the instruction execution outside all or part of the state, or controlled the outside world. It can still change the outside world at different degree to simulate real outside all sorts of different environment’s impact on the region.3.2 All kinds of Agent ModelingBased on MAS model, the Agent is behavior corpus, on certain condition, it can response to some external events, namely in some activities, making the state transition or produce new events. Agent has level-oriented, simultaneity, it can use triad description: <attribute,structure, behavior>. For the ecological complex adaptive system model of the Agent, attributes can be divided into three categories. They are the space attributes, physical properties, and the other attributes. Structure refers to the internal Agent contained hypo-Agent or entity set, the structure of the hypo-Agent or entity attributes, and the Agents attributes of its internal state jointly characterized. Agent actions are divided into reactive behavior and adaptability behavior. Reactive behavior refers tothe Agent on its environment conditions, internal state and other driver response. Adaptive behavior refers to the Agent from environmental and interaction with other Agent in the process of learning and accumulate experience, and improve their adaptability to fit the environment activities. This model Agent in the main internal structure is a reactive Agent. In order to stimulate the way, a response to external the environment is needed to respond to changes, thus realize the ecological system evolution and propulsion.1. Inorganic environmental Agent: through different environments, set the intensity of photosynthesis, the producers and the fitness degree of the consumers Agents, to achieve a real return based on the external environment.2. Producers Agent: First of all to acquire the behavior of inorganic environment of the Agent to carry out their actions. It has some static properties and it does not change in Agent's life cycle. Producer Agent's static properties: self-energy (Energy), growth time, position (x, y), color (Color). Producer Agent behavior: growth, according to the conditions of inorganic environment to observe whether theycan growth and the growth rate.3. The Consumer Agent: definite the two kinds of consumers Agents, herbivorous Consumer Agent and carnivorous consumers Agent. The Herbivorous consumers eat producers Agent to get energy and the carnivorous Agent eats herbivorous Agent to get energy. They have available activities on certain regional, and if the mortality of consumers Agent excess in this region so inorganic environment by Agent producers will increase the impact on growth.The consumers Agent static properties: Energy, Reproduce, Energy consume, Gain, Energy max, Range. Consumer Agent’s behavior:(1) Foraging rules: the consumer Agent do activities within certain regions, they find food through close interaction with each other but were different Agent. The herbivorous Agent and Producer Agent interact with each other for feeding. The carnivorous Agent and herbivorous Agent interact with each other for food.(2) Breeding rules: the consumer Agent through the reproduction rate to breeding which set by inorganic environment, the inorganic environment of Agent through reality conditions to set corresponding reproduction rate.(3) Death rules: all kinds of Agent’s energy less than0 or be established, then death.4 Model Simulation4.1 Many Agent Simulation Platform NetLogoNet Logo is a multi-Agent modeling and simulation integrated environment, especially suitable for the modeling and simulation of complex systems with time evolution. Net Logo is developed by the Northwestern University to connect Center for Connected Learning and Computer-Based Modeling, CCL. The objective is to provide powerful CAD tools for research and education [8]. Net Logo modeling assumptions: the space is divided into grid, each grid is a static Agent, the Mobile Agents located in two-dimensional space, each Agent have independent action, all of the main body doing parallel and asynchronous update, the entire system as time and dynamic changes. Net Logo defines the three types of Agent: territory, Turtle, Observer. The Patch which is static Agent, it divided our virtual space into grids, Turtle is a dynamic Agent, it can also be a variety, it is moving in the Patch randomly,you can interact with each other and interacted with Patch. The Observer as a global Agent observed the local world which composited by the Turtle and the Patch, and control the environment.4.2 Simulation Model and Results Analysis4.2.1 Simulation 1Inorganic environment ecosystem model requires relatively abundant photosynthesis and water conditions. First set all kinds of Agent in photosynthesis and enough moisture case attributes. Plant Agent’s properties:(1) Obtain photosynthesis fully.(2) Plant Agent’s growth rate: Grass growth = 9animals Agent’s properties.(3) Number of initial population: Initial Sheep Number = 99, Initial Wolf Number = 49(4) Its own energy: Initial Energy randomlygenerated between 1 and 4.(5) The energy consumption: each move of Energy Consume = 1.(6) Range: Range = 3.(7) Access to energy: Sheep Gain = 5, Wolf Gain = 15.(8) Reproduction rate: Sheep Reproduce = 3%, Wolf Reproduce = 5%.4.2.2 Simulation 2Put environment into water is not adequate, then the survival rate of plants and biological become low. Other conditions are the same as Simulation 1, only put the environment into arid environment, photosynthesis and water is not sufficient, plant growth rate become low.4.2.3 Simulation 3Conditions and simulation experiments 2 are identical, increasing the fitness of the biologicalcommunity to the environment.(1) The access to energy: Sheep Gain = 8, Wolf Gain = 30.(2) The reproductive rate: Sheep Reproduce = 6%,Wolf Reproduce = 10%. From the above simulations show, when the population increased the fitness for environment, the fluctuations is high in various types of population at the beginning, and then have some down, but gradually adapt to the environment, the consumers population are becoming more and more balanced. This is show that the poor ecological environments are not suitable for the survival of biological communities. Some species will be eliminated, if it cannot adapt to the biological environment, and the number will decline. However, when the populations adapt to the environment, the population size will gradually stabilized, will not lead to extinction of biological communities in a population, over time, will achieve the ecological balance. This feature is fully consistent with the natural survival of the fittest, the law of survival of the fittest.5 ConclusionThis paper based on the theory of complex adaptive systems, through a multi-Agent modeling method, to establish a grassland ecosystem model based on reaction type Agent. The model thought the properties and behavior of various types of Agent to feedback and to complete the entire eco-system evolution and progress. Then in the Net Logo simulation platform, by changing the environmental factors andattribute parameters of various types of Agent which observed the changes in population, especially the population which has high adaptability in harsh environments, the low fitness wereeliminated or even be extinct, reflecting the characteristics of an ecosystem. Extension of the model is high, in the case of environmental change, just by different populations of Agent to change the properties and behavior can be simulated different ecosystems, such as marine ecosystems. However, the text of the Agent only defines the behavior of reactive, not adaptive, so I hope you can join in the follow-up to all kinds of Agent's adaptive behavior to improve the system and make it to become a more realistic ecosystem, ultimately to provide services for the interaction between humans and the environment and ecological harmony.AcknowledgementsThis paper is supported by the National Natural Science Foundation of China (40872087) and the Shaanxi Provincial Key Subjects (Computer Application Technology).References:[1] Jiao Licheng, Liu Jing, Zhong Wecai and so on. Coevolutionary Computation and Multi-agent System. Beijing: Science Press. 2006, 34 ~ 51.[2] Shi Zhongzhi. Intelligent Agent and Its Application[M]. Beijing: Science Press. 2000.[3] Yu Jiangtao. The research and application of Multi-agent model, learning, and collaborative[PhD thesis]. Hangzhou: Zhejiang University.2003.[4] Potter MA ， De Jong KA. Cooperative coevolution: an architecture for evolving coadapted subcomponents. Evolutionary Compution. 2000, 8(1):1~29.[5] WEI Yun,HAN Yin,Fan Bing-quan. Modeling and simulation of intersection based on multi-agents and fuzzy control strategy. Journal of University of Shanghai for Science and Technology. 2010, Vol. 32, No. 3, 259~262.[6] Jian-Chao Zeng, Hong-Gang Wang. Modeling and Simulation of the food Chain Based on Complex Adaptive System. 2009.[7] Cheng J,Liu H,Zhang H.Primary study on MAS-based single-species ecosystem model. SPCA 2006: 2006 1st International Symposium on Pervasive Computing and Applications. United States: Computer Society, 2007.专业英语课程大作业题目：Modeling and Simulation of EcosystemBased on Multi-Agent System院系：电子信息与电气工程学院学生姓名：杨锐学号：201002070007专业班级：电子信息工程（10级专升本）2011年10月22 日。

[S3600]H3C交换机配置详解本文来源于网络，感觉比较全面，给大家分享；一.用户配置:<H3C>system-view[H3C]super password H3C 设置用户分级密码[H3C]undo super password 删除用户分级密码[H3C]localuser bigheap 123456 1 Web网管用户设置,1（缺省）为管理级用户,缺省admin,admin [H3C]undo localuser bigheap 删除Web网管用户[H3C]user-interface aux 0 只支持0[H3C-Aux]idle-timeout 2 50 设置超时为2分50秒,若为0则表示不超时,默认为5分钟[H3C-Aux]undo idle-timeout 恢复默认值[H3C]user-interface vty 0 只支持0和1[H3C-vty]idle-timeout 2 50 设置超时为2分50秒,若为0则表示不超时,默认为5分钟[H3C-vty]undo idle-timeout 恢复默认值[H3C-vty]set authentication password 123456 设置telnet密码,必须设置[H3C-vty]undo set authentication password 取消密码[H3C]display users 显示用户[H3C]display user-interface 显示用户界面状态二.系统IP配置:[H3C]vlan 20[H3C]management-vlan 20[H3C]interface vlan-interface 20 创建并进入管理VLAN[H3C]undo interface vlan-interface 20 删除管理VLAN接口[H3C-Vlan-interface20]ip address 192.168.1.2 255.255.255.0 配置管理VLAN接口静态IP地址(缺省为192.168.0.234)[H3C-Vlan-interface20]undo ip address 删除IP地址[H3C-Vlan-interface20]ip gateway 192.168.1.1 指定缺省网关(默认无网关地址)[H3C-Vlan-interface20]undo ip gateway[H3C-Vlan-interface20]shutdown 关闭接口[H3C-Vlan-interface20]undo shutdown 开启[H3C]display ip 显示管理VLAN接口IP的相关信息[H3C]display interface vlan-interface 20 查看管理VLAN的接口信息<H3C>debugging ip 开启IP调试功能<H3C>undo debugging ip三.DHCP客户端配置:[H3C-Vlan-interface20]ip address dhcp-alloc 管理VLAN接口通过DHCP方式获取IP地址[H3C-Vlan-interface20]undo ip address dhcp-alloc 取消[H3C]display dhcp 显示DHCP客户信息<H3C>debugging dhcp-alloc 开启DHCP调试功能<H3C>undo debugging dhcp-alloc四.端口配置:[H3C]interface Ethernet0/3[H3C-Ethernet0/3]shutdown[H3C-Ethernet0/3]speed 100 速率,可为10,100,1000和auto(缺省)[H3C-Ethernet0/3]duplex full 双工,可为half,full和auto(缺省) 光口和汇聚后不能配置[H3C-Ethernet0/3]flow-control 开启流控,默认为关闭[H3C-Ethernet0/3]broadcast-suppression 20 设置抑制广播百分比为20%,可取5,10,20,100,缺省为100,同时组播和未知单播也受此影响[H3C-Ethernet0/3]loopback internal 内环测试[H3C-Ethernet0/3]loopback external 外环测试,需插接自环头,必须为全双工或者自协商模式[H3C-Ethernet0/3]port link-type trunk 设置链路的类型为trunk,可为access(缺省),trunk[H3C-Ethernet0/3]port trunk pvid vlan 20 设置20为该trunk的缺省VLAN,默认为1(trunk线路两端的PVID必须一致)[H3C-Ethernet0/3]port access vlan 20 将当前access端口加入指定的VLAN[H3C-Ethernet0/3]port trunk permit vlan all 允许所有的VLAN通过当前的trunk端口,可多次使用该命令[H3C-Ethernet0/3]mdi auto 设置以太端口为自动监测,normal(缺省)为直通线,across为交叉线[H3C]link-aggregation Ethernet 0/1 to Ethernet 0/4 将1-4口加入汇聚组,1为主端口,两端需要同时配置,设置了端口镜像以及端口隔离的端口无法汇聚[H3C]undo link-aggregation Ethernet 0/1 删除该汇聚组[H3C]link-aggregation mode egress 配置端口汇聚模式为根据目的MAC地址进行负荷分担,可选为ingress,egress和both,缺省为both[H3C]monitor-port Ethernet 0/2 将该端口设置为镜像端口,必须先设置镜像端口,删除时必须先删除被镜像端口,而且它们不能同在一个端口,该端口不能在汇聚组中,设置新镜像端口时,新取代旧,被镜像不变[H3C]mirroring-port Ethernet 0/3 to Ethernet 0/4 both 将端口3和4设置为被镜像端口,both为同时监控接收和发送的报文,inbound表示仅监控接收的报文,outbound表示仅监控发送的报文[H3C]display mirror[H3C]display interface Ethernet 0/3< H3C>reset counters 清除所有端口的统计信息[H3C]display link-aggregation Ethernet 0/3 显示端口汇聚信息[H3C-Ethernet0/3]virtual-cable-test 诊断该端口的电路状况五.VLAN配置:[H3C]vlan 2[H3C]undo vlan all 删除除缺省VLAN外的所有VLAN,缺省VLAN不能被删除[H3C-vlan2]port Ethernet 0/4 to Ethernet 0/7 将4到7号端口加入到VLAN2中,此命令只能用来加access端口,不能用来增加trunk或者hybrid端口[H3C-vlan2]port-isolate enable 打开VLAN内端口隔离特性,不能二层转发,默认不启用该功能[H3C-Ethernet0/4]port-isolate uplink-port vlan 2 设置4为VLAN2的隔离上行端口,用于转发二层数据,只能配置一个上行端口,若为trunk,则建议允许所有VLAN通过,隔离不能与汇聚同时配置[H3C]display vlan all 显示所有VLAN的详细信息S1550E支持基于端口的VLAN,通过创建不同的user-group来实现,一个端口可以属于多个user-group,不属于同一个user-group的端口不能互相通信,最多支持50个user-group[H3C]user-group 20 创建user-group 20,默认只存在user-group 1[H3C-UserGroup20]port Ethernet 0/4 to Ethernet 0/7 将4到7号端口加入到VLAN20中,初始时都属于user-group 1中[H3C]display user-group 20 显示user-group 20的相关信息六.集群配置:S2100只能作为成员交换机加入集群中,加入后系统名改为"集群名_成员编号.原系统名"的格式.即插即用功能通过两个功能实现: 集群管理协议MAC组播地址协商和管理VLAN协商[H3C]cluster enable 启用群集功能,缺省为启用[H3C]cluster 进入群集视图[H3C-cluster]administrator-address H-H-H name switch H-H-H为命令交换机的MAC,加入switch集群[switch_1.H3C-cluster]undo administrator-address 退出集群[H3C]display cluster 显示集群信息[H3C]management-vlan 2 集群报文只能在管理VLAN中转发,同一集群需在同一个管理VLAN中,需在建立集群之前指定管理VLAN< H3C>debugging cluster七.QoS配置:QoS配置步骤:设置端口的优先级,设置交换机信任报文的优先级方式,队列调度,端口限速[H3C-Ethernet0/3]priority 7 设置端口优先级为7,默认为0[H3C]priority-trust cos 设置交换机信任报文的优先级方式为cos(802.1p优先级,缺省值),还可以设为dscp方式(dscp优先级方式)[H3C]queue-scheduler hq-wrr 2 4 6 8 设置队列调度算法为HQ-WRR(默认为WRR),权重为2,4,6,8[H3C-Ethernet0/3]line-rate inbound 29 将端口进口速率限制为2Mbps,取1-28时,速率为rate*8*1024/125,即64,128,192...1.792M;29-127时,速率为(rate-27)*1024,即2M,3M,4M...100M,千兆时可继续往下取,128-240时,速率为(rate-115)*8*1024,即104M,112M,120M...1000M [H3C]display queue-scheduler 显示队列调度模式及参数[H3C]display priority-trust 显示优先级信任模式八.系统管理:[H3C]mac-address blackhole H-H-H vlan 1 在VLAN1中添加黑洞MAC[H3C]mac-address static H-H-H interface Ethernet 0/1 vlan 1 在VLAN1中添加端口一的一个mac[H3C]mac-address timer aging 500 设置MAC地址表的老化时间为500s[H3C]display mac-address[H3C]display arp[H3C]mac-address port-binding H-H-H interface Ethernet 0/1 vlan 1 配置端口邦定[H3C]display mac-address port-binding[H3C]display saved-configuration[H3C]display current-configuration< H3C>save[H3C]restore default 恢复交换机出厂默认配置,恢复后需重启才能生效[H3C]display version< H3C>reboot[H3C]display device[H3C]sysname bigheap[H3C]info-center enable 启用系统日志功能,缺省情况下启用[H3C]info-center loghost ip 192.168.0.3 向指定日志主机(只能为UNIX或LINUX,不能为Windows)输出信息,需先开启日志功能,缺省关闭[H3C]info-center loghost level 8 设置系统日志级别为8,默认为5.级别说明:1.emergencies 2.alerts 3.critical 4.errors 5.warnings 6.notifications rmational 8.debugging< H3C>terminal debugging 启用控制台对调试信息的显示,缺省控制台为禁用<H3C>terminal logging 启用控制台对日志信息的显示,缺省控制台为启用<H3C>terminal trapping 启用控制台对告警信息的显示,缺省控制台为启用[H3C]display info-center 显示系统日志的配置和缓冲区记录的信息[H3C]display logbuffer 显示日志缓冲区最近记录的指定数目的日志信息[H3C]display trapbuffer 显示告警缓冲区最近记录的指定数目的日志信息<H3C>reset logbuffer 清除日志缓冲区的信息<H3C>reset trapbuffer 清除告警缓冲区的信息九.网络协议配置:NDP即是邻居发现协议,S1550E只能开启或关闭NDP,无法配置,默认有效保留时间为180s,NDP报文发送的间隔60s[H3C]ndp enable 缺省情况下是开启的[H3C-Ethernet0/3]ndp enable 缺省情况下开启[H3C]display ndp 显示NDP配置信息[H3C]display ndp interface Ethernet 0/1 显示指定端口NDP发现的邻居信息<H3C>debugging ndp interface Ethernet 0/1HABP协议即Huawei Authentication Bypass Protocol,华为鉴权旁路协议,是用来解决当交换机上同时配置了802.1x和HGMPv1/v2时,未经授权和认证的端口上将过滤HGMP报文,从而使管理设备无法管理下挂的交换机的问题。