Алг формир рекомендац при планир инф обмена ... Балашов (2007) (Лекции 2014-2015), страница 4

1.2. Å˚ÎË Á‡ÙËÍÒËÓ‚‡Ì˚ Á̇˜ÂÌËfl ÒÎÂ‰Û˛˘Ëı Ú·ӂ‡ÌËÈ ÍËÌÙÓχˆËÓÌÌÓÏÛ Ó·ÏÂÌÛ:‰ÎË̇ ÔÓ‰ˆËÍ·: 2000 ÏÍÒ;2 í‡ÍÚ͇̇· åäàé ËÏÂÂÚ ‰ÎËÚÂθÌÓÒÚ¸ 1 ÏÍÒ.àáÇÖëíàü êÄç. íÖéêàü à ëàëíÖåõ ìèêÄÇãÖçàü      ‹ 6      2007ÄãÉéêàíåõ îéêåàêéÇÄçàü êÖäéåÖçÑÄñàâ èêà èãÄçàêéÇÄçàà퇷Îˈ‡ 1. á̇˜ÂÌËfl ËÁÏÂÌflÂÏ˚ı Ú·ӂ‡ÌËÈ Í ËÌÙÓχˆËÓÌÌÓÏÛ Ó·ÏÂÌÛ燘‡Î¸ÌÓÂÁ̇˜ÂÌËÂ, initiÉ‡Ì˘ÌÓÂÁ̇˜ÂÌËÂ, limiRF0.50MPE0.011MCC1í·ӂ‡ÌËÂ, Ri15χÍÒËχθ̇fl ‰ÎËÚÂθÌÓÒÚ¸ ˆÂÔÓ˜ÍË ‡·ÓÚ:2000 ÏÍÒ;˜‡ÒÚÓÚ˚ Ë ‰ÎËÚÂθÌÓÒÚË ÒÓÓ·˘ÂÌËÈ.á̇˜ÂÌËfl ÒÎÂ‰Û˛˘Ëı Ú·ӂ‡ÌËÈ ·˚ÎÓ ‰ÓÔÛÒÚËÏÓ ËÁÏÂÌflÚ¸:1) Á‡ÂÁÂ‚ËÓ‚‡Ì̇fl ‰ÓÎfl ‰ÎËÌ˚ ÔÓ‰ˆËÍ·(Ó·ÓÁ̇˜ÂÌËÂ: RF, ÓÚ reserved fraction);2) χÍÒËχθÌÓ ‰ÓÔÛÒÚËÏÓ ÓÚÍÎÓÌÂÌË ‡ÒÒÚÓflÌËfl ÏÂÊ‰Û ‡·ÓÚ‡ÏË Ó‰ÌÓ„Ó Ë ÚÓ„Ó Ê ÒÓÓ·˘ÂÌËfl ÓÚ ÔÂËÓ‰‡ ˝ÚÓ„Ó ÒÓÓ·˘ÂÌËfl (Ó·ÓÁ̇˜ÂÌËÂ:MPE, ÓÚ maximum period error);3) χÍÒËχθÌÓ ‰ÓÔÛÒÚËÏÓ ˜ËÒÎÓ ‡·ÓÚ ‚ ˆÂÔӘ͠‡·ÓÚ (Ó·ÓÁ̇˜ÂÌËÂ: MCJ, ÓÚ maximum[number of] chained jobs).燘‡Î¸Ì˚ (ÌÂÒÓ‚ÏÂÒÚËÏ˚Â) Á̇˜ÂÌËfl Ú·ӂ‡ÌËÈ 1)–3) Ë „‡Ìˈ˚ ËÁÏÂÌÂÌËÈ ˝ÚËı Ú·ӂ‡ÌËÈ ·˚ÎË ‚˚·‡Ì˚ ÒÓÓÚ‚ÂÚÒÚ‚ÂÌÌÓ Ú‡·Î.

1 Ú‡Í, ˜ÚÓ·˚ Á̇˜ÂÌËfl Ú·ӂ‡ÌËÈ ËÁ ̇·Ó‡ LIM ·˚ÎË ÒÓ‚ÏÂÒÚËÏ˚ÏË, ‡ Á̇˜ÂÌËfl ËÁ ̇·Ó‡ INIT – ÌÂÒÓ‚ÏÂÒÚËÏ˚ÏË.è‡‡ÏÂÚ dcost ‡Î„ÓËÚχ ·˚Î ÔËÌflÚ ‡‚Ì˚Ï 0.1.Å˚ÎË Ôӂ‰ÂÌ˚ ˝ÍÒÔÂËÏÂÌÚ˚ Ò ‡Á΢Ì˚ÏË Ì‡·Ó‡ÏË ‚ÂÒÓ‚˚ı ÍÓ˝ÙÙˈËÂÌÚÓ‚ creli (ÒÏ.‡Á‰. 2.1). ÑÎfl Í‡Ê‰Ó„Ó Ì‡·Ó‡ ÍÓ˝ÙÙˈËÂÌÚÓ‚ÏÂÚÓ‰ÓÏ ÔÓÎÌÓ„Ó ÔÂ·Ó‡ ÓÔ‰ÂÎflÎÓÒ¸ ÓÔÚËχθÌÓ (Ò ÏËÌËχθÌ˚Ï Á̇˜ÂÌËÂÏ ˆÂ΂ÓÈ ÙÛÌÍ-83ˆËË Cost) ¯ÂÌË Xopt Á‡‰‡˜Ë ÙÓÏËÓ‚‡ÌËfl ÂÍÓÏẨ‡ˆËÈ. èÓ ÂÁÛÎ¸Ú‡Ú‡Ï Í‡Ê‰Ó„Ó ˝ÍÒÔÂËÏÂÌÚ‡·˚· ‚˚˜ËÒÎÂ̇ ‡ÁÌÓÒÚ¸ Á̇˜ÂÌËÈ ˆÂ΂ÓÈ ÙÛÌ͈ËË Cost ̇ ÓÔÚËχθÌÓÏ ¯ÂÌËË Xopt Ë ¯ÂÌËËXres, ÔÓÎÛ˜ÂÌÌÓÏ ÓÚ ËÒÒΉÛÂÏÓ„Ó ‡Î„ÓËÚχ.êÂÁÛθڇÚ˚ ˝ÍÒÔÂËÏÂÌÚÓ‚ Ô˂‰ÂÌ˚ ‚ Ú‡·Î. 2.äÛÒË‚ÓÏ ‚˚‰ÂÎÂÌ˚ Á̇˜ÂÌËfl Ú·ӂ‡ÌËÈ ËÁ Xres,ÓÚ΢Ì˚ ÓÚ Á̇˜ÂÌËÈ ËÁ Xopt. êÂÁÛθڇÚ˚ Ôӂ‰ÂÌÌ˚ı ˝ÍÒÔÂËÏÂÌÚÓ‚ ÔÓ͇Á˚‚‡˛Ú, ˜ÚÓ Ì‡ ‚˚·‡ÌÌÓÏÔËÏÂ ÓÚÍÎÓÌÂÌË Á̇˜ÂÌËfl ˆÂ΂ÓÈ ÙÛÌ͈ËË Ì‡ÂÁÛθڇÚ ‡·ÓÚ˚ ‡Î„ÓËÚχ ÓÚ ÓÔÚËχθÌÓ„Ó Á̇˜ÂÌËfl ˆÂ΂ÓÈ ÙÛÌ͈ËË Ì Ô‚˚¯‡ÂÚ 2% ÓÚ ÓÔÚËχθÌÓ„Ó Á̇˜ÂÌËfl.

èË ˝ÚÓÏ ‰Îfl ËÁÏÂÌflÂÏÓ„Ó Ú·ӂ‡ÌËfl Ò Ï‡ÍÒËχθÌ˚Ï ‚ÂÒÓ‚˚Ï ÍÓ˝ÙÙˈËÂÌÚÓχ΄ÓËÚÏ ‚˚‰‡ÂÚ Á̇˜ÂÌËÂ, ÒÓ‚Ô‡‰‡˛˘Â ÒÓ Á̇˜ÂÌËÂÏ ËÁ ÓÔÚËχθÌÓ„Ó ¯ÂÌËfl. çÂÓ·ıÓ‰ËÏÓ ÓÚÏÂÚËÚ¸, ˜ÚÓ ‚ ˝ÍÒÔÂËÏÂÌÚ ‹ 3 ‚˚˜ËÒÎÂÌÌÓ ‡Î„ÓËÚÏÓÏ Á̇˜ÂÌË MPE ‚ 2 ‡Á‡ ÓÚ΢‡ÂÚÒfl ÓÚ ÓÔÚËχθÌÓ„Ó. ùÚÓ „Ó‚ÓËÚ Ó ˆÂÎÂÒÓÓ·‡ÁÌÓÒÚËÔËÏÂÌÂÌËfl Í ÔÓÎÛ˜ÂÌÌÓÏÛ ¯ÂÌ˲ Ôӈ‰Û˚,‚˚ÔÓÎÌfl˛˘ÂÈ ÛÒËÎÂÌË ÓÚ‰ÂθÌ˚ı ËÁÎ˯Ì ÓÒ··ÎÂÌÌ˚ı Ú·ӂ‡ÌËÈ. ùÚÓ ÏÓÊÂÚ ·˚Ú¸, ̇ÔËÏÂ,Ôӈ‰Û‡ refine_compatible, ÓÔËÒ‡Ì̇fl ‚ ‡Á‰.

3.2.èË ËÒÒΉӂ‡ÌËË ‡Î„ÓËÚχ ̇ ËÒÍÛÒÒÚ‚ÂÌÌÓÒÙÓÏËÓ‚‡ÌÌ˚ı ̇·Ó‡ı ÒÓÓ·˘ÂÌËÈ (‚ ˜‡ÒÚÌÓÒÚË,Ô˂‰ÂÌÌ˚ı ‚ [13]) ·˚ÎÓ ‚˚fl‚ÎÂÌÓ ÓÚÍÎÓÌÂÌË ÓÚÓÔÚËχθÌÓ„Ó ¯ÂÌËfl ‰Ó 10% ÔÓ Á̇˜ÂÌ˲ ˆÂ΂ÓÈ ÙÛÌ͈ËË Cost. èË Û‚Â΢ÂÌËË ˜ËÒ· ËÁÏÂÌflÂÏ˚ı ÚÂıÌÓÎӄ˘ÂÒÍËı Ó„‡Ì˘ÂÌËÈ ÓÚÍÎÓÌÂÌË ÒÚ‡ÎÓ ‰ÓÒÚË„‡Ú¸ 15–20%. ùÚÓ ÔÓÁ‚ÓÎflÂÚ Ò‰Â·ڸ ‚˚‚Ó‰Ó ÌÂÓ·ıÓ‰ËÏÓÒÚË ÔËÏÂÌÂÌËfl ‡Î„ÓËÚÏÓ‚, ‚˚ÔÓÎÌfl˛˘Ëı ÔÓÒΉӂ‡ÚÂθÌÓ ÛÏÂ̸¯ÂÌË Á̇˜ÂÌËflˆÂ΂ÓÈ ÙÛÌ͈ËË Ì‡ ÚÂÍÛ˘ÂÏ ÔË·ÎËÊÂÌËË, Ú‡ÍËı,Í‡Í Ô‰ÎÓÊÂÌÌ˚È ‚ ‡Á‰.

3.2 ‡Î„ÓËÚÏ Ò ÔÓÒΉӂ‡ÚÂθÌ˚Ï ÒÛÊÂÌËÂÏ Ó·Î‡ÒÚË ÔÓËÒ͇. Ç Í‡˜ÂÒڂ ̇˜‡Î¸ÌÓ„Ó ÔË·ÎËÊÂÌËfl ‰Îfl Ú‡ÍËı ‡Î„ÓËÚÏÓ‚ ÏÓ-퇷Îˈ‡ 2. êÂÁÛθڇÚ˚ ˝ÍÒÔÂËÏÂÌÚÓ‚çÓÏÂ˝ÍÒÔÂËÏÂÌÚ‡123RiCRELiresoptxixiRF100.330.33MPE100.090.09MCC1033RF100.330.33MPE1000.080.08MCC1RF11Cost(Xres) – Cost(Xopt)0Cost(Xopt)5.630.0711.110.1611.0810100.310.31MPE10.310.15MCC10022àáÇÖëíàü êÄç. íÖéêàü à ëàëíÖåõ ìèêÄÇãÖçàü      ‹ 6      20076*Ň·¯Ó‚84ÊÂÚ ‚˚ÒÚÛÔ‡Ú¸ ¯ÂÌËÂ, ÔÓÎÛ˜ÂÌÌÓ ‡Î„ÓËÚÏÓÏ Ò˝‚ËÒÚ˘ÂÒÍËÏ ‚˚·ÓÓÏ ÓÒ··ÎflÂÏÓ„Ó Ú·ӂ‡ÌËfl.á‡Íβ˜ÂÌËÂ. è‰ÎÓÊÂÌ˚ ‡Î„ÓËÚÏ˚ ÙÓÏËÓ‚‡ÌËfl ÂÍÓÏẨ‡ˆËÈ ÔÓ ËÁÏÂÌÂÌ˲ ÌÂÒÓ‚ÏÂÒÚËÏ˚ı Ú·ӂ‡ÌËÈ Í ËÌÙÓχˆËÓÌÌÓÏÛ Ó·ÏÂÌÛÔÓ Í‡Ì‡ÎÛ Ò ˆÂÌÚ‡ÎËÁÓ‚‡ÌÌ˚Ï ÛÔ‡‚ÎÂÌËÂÏ ‰ÎflÓ·ÂÒÔ˜ÂÌËfl Ëı ÒÓ‚ÏÂÒÚËÏÓÒÚË.

è˂‰ÂÌ˚ ÂÁÛθڇÚ˚ ˝ÍÒÔÂËÏÂÌڇθÌÓ„Ó ËÒÒΉӂ‡ÌËfl Ó‰ÌÓ„Ó ËÁ ‡Î„ÓËÚÏÓ‚ ̇ ‰‡ÌÌ˚ı ÔÓ ‡θÌÓÈ ÇëêÇ.àÒÒΉӂ‡ÌË ÔÓ͇Á‡ÎÓ ‚˚ÒÓÍÛ˛ ˝ÙÙÂÍÚË‚ÌÓÒÚ¸‡Î„ÓËÚχ ÔÓ ÍËÚÂ˲ ÏËÌËχθÌÓ„Ó ÒÛÏχÌÓ„Ó ËÁÏÂÌÂÌËfl Á̇˜ÂÌËÈ ËÒıÓ‰Ì˚ı Ú·ӂ‡ÌËÈ.Ä΄ÓËÚÏ Ò ˝‚ËÒÚ˘ÂÒÍËÏ ‚˚·ÓÓÏ ËÁÏÂÌflÂÏÓ„Ó Ú·ӂ‡ÌËfl ÔËÏÂÌflÂÚÒfl ‚ ÒÓÒÚ‡‚ ËÌÒÚÛÏÂÌڇθÌÓÈ ÒËÒÚÂÏ˚ ‡‚ÚÓχÚ˘ÂÒÍÓ„Ó Ô·ÌËÓ‚‡ÌËfl ËÌÙÓχˆËÓÌÌÓ„Ó Ó·ÏÂ̇ ÔÓ Í‡Ì‡ÎÛ Ò ˆÂÌÚ‡ÎËÁÓ‚‡ÌÌ˚Ï ÛÔ‡‚ÎÂÌËÂÏ [3, 4], ËÒÔÓθÁÛÂÏÓÈ ‚ÔÓÏ˚¯ÎÂÌÌÓÒÚË ÔË ÔÓÂÍÚËÓ‚‡ÌËË ·ÓÚÓ‚˚ıÇëêÇ ‰Îfl ‡‚ˇˆËÓÌÌ˚ı Ë ÏÓÒÍËı ÍÓÏÔÎÂÍÒÓ‚.ëÔËÒÓÍ ÎËÚÂ‡ÚÛ˚1. ÉÓÒÛ‰‡ÒÚ‚ÂÌÌ˚È Òڇ̉‡Ú êî “àÌÚÂÙÂÈÒ Ï‡„ËÒÚ‡Î¸Ì˚È ÔÓÒΉӂ‡ÚÂθÌ˚È ÒËÒÚÂÏ˚ ˝ÎÂÍÚÓÌÌ˚È ÏÓ‰ÛÎÂÈ” Ééëí ê 52070-2003.2.

äÓÒÚÂÌÍÓ Ç.Ä., ÉÛ¸flÌÓ‚ Ö.ë. Ä΄ÓËÚÏ ÔÓÒÚÓÂÌËfl ‡ÒÔËÒ‡ÌËÈ Ó·ÏÂÌÓ‚ ÔÓ ¯ËÌÂ Ò ˆÂÌÚ‡ÎËÁÓ‚‡ÌÌ˚Ï ÛÔ‡‚ÎÂÌËÂÏ Ë ËÒÒΉӂ‡ÌËÂ Â„Ó ˝ÙÙÂÍÚË‚ÌÓÒÚË // èÓ„‡ÏÏËÓ‚‡ÌËÂ. 2005. ‹ 6. ë. 340–346.3. Ň·¯Ó‚ Ç.Ç., LJ‚ËÌÓ‚ ë.Ç., ÉÛ¸flÌÓ‚ Ö.ë. Ë ‰.ëËÒÚÂχ ‡‚ÚÓχÚ˘ÂÒÍÓ„Ó ÔÓÒÚÓÂÌËfl ˆËÍÎÓ„‡ÏÏ˚ Ó·ÏÂÌÓ‚ ÔÓ ¯ËÌÂ Ò ˆÂÌÚ‡ÎËÁÓ‚‡ÌÌ˚Ï ÛÔ‡‚ÎÂÌËÂÏ // í. ‚ÚÓÓÈ ÇÒÂÓÒÒËÈÒÍ. ̇ۘÌ. ÍÓÌÙ.“åÂÚÓ‰˚ Ë Ò‰ÒÚ‚‡ Ó·‡·ÓÚÍË ËÌÙÓχˆËË”. å.:àÁ‰‡ÚÂθÒÍËÈ ÓÚ‰ÂÎ Ù‡ÍÛθÚÂÚ‡ ‚˚˜ËÒÎËÚÂθÌÓÈχÚÂχÚËÍË Ë ÍË·ÂÌÂÚËÍË åÉì, 2005.

ë. 516–521.4. Balashov V.V., Kostenko V.A., Smeliansky R.L. et al. ATool System for Automatic Scheduling of Data Exchange in Real-Time Distributed Embedded Systems //Proc. Seventh IEEE Internat. Sympos. on Computer Networks (ISCN'06). Istanbul, Turkey, 2006.5. óËÒÚÓÎËÌÓ‚ å.Ç. åÂÚÓ‰˚ Ë Ò‰ÒÚ‚‡ ÙÛÌ͈ËÓ̇θÌÓ„Ó ÚÂÒÚËÓ‚‡ÌËfl ÅñÇå-àÇë // í. ‚ÚÓÓÈÇÒÂÓÒÒËÈÒÍ. ̇ۘÌ.

ÍÓÌÙ. “åÂÚÓ‰˚ Ë Ò‰ÒÚ‚‡ Ó·‡·ÓÚÍË ËÌÙÓχˆËË”. å.: àÁ‰‡ÚÂθÒÍËÈ ÓÚ‰ÂÎÙ‡ÍÛθÚÂÚ‡ ‚˚˜ËÒÎËÚÂθÌÓÈ Ï‡ÚÂχÚËÍË Ë ÍË·ÂÌÂÚËÍË åÉì, 2005. ë. 442–448.6. åËÌÛ å. å‡ÚÂχÚ˘ÂÒÍÓ ÔÓ„‡ÏÏËÓ‚‡ÌËÂ.íÂÓËfl Ë ‡Î„ÓËÚÏ˚. å.: ç‡Û͇, 1990.7. Ň·¯Ó‚ Ç.Ç. èÓ‰‰ÂÊ͇ ÔËÌflÚËfl ¯ÂÌËÈ ÔËÔÓÒÚÓÂÌËË ˆËÍÎÓ„‡ÏÏ˚ Ó·ÏÂÌÓ‚ ÔÓ ÏÛθÚËÔÎÂÍÒÌÓÏÛ Í‡Ì‡ÎÛ // í. ‚ÚÓÓÈ ÇÒÂÓÒÒËÈÒÍ. ̇ۘÌ.

ÍÓÌÙ. “åÂÚÓ‰˚ Ë Ò‰ÒÚ‚‡ Ó·‡·ÓÚÍË ËÌÙÓχˆËË”. å.: àÁ‰‡ÚÂθÒÍËÈ ÓÚ‰ÂÎ Ù‡ÍÛθÚÂÚ‡ ‚˚˜ËÒÎËÚÂθÌÓÈ Ï‡ÚÂχÚËÍË Ë ÍË·ÂÌÂÚËÍË åÉì,2005. ë. 508–515.8. Bate I.J. Scheduling and Timing Analysis for SafetyCritical Real-Time Systems // Department of ComputerScience. University of York, 1999.9. àÁχÈÎÓ‚ Ä.î., ëÓÎÓ‰Ó‚ å.Ç. óËÒÎÂÌÌ˚ ÏÂÚÓ‰˚ÓÔÚËÏËÁ‡ˆËË. å.: îËÁχÚÎËÚ, 2005.10.

Stewart D.B., Arora G. A Tool for Analyzing and FineTuning the Real-Time Properties of an Embedded System // IEEE Trans. on Software Engineering. Issue 4.2003. V. 29.11. Park J., Ryu M., Hong S. et al. Rapid performance re-engineering of distributed embedded systems via latencyanalysis and k-level diagonal search // J. of Parallel andDistributed Computing. Issue 1.

2006. V. 66.12. Pop T., Eles P., Peng Z. Schedulability Analysis for Distributed Heterogeneous Time / Event Triggered RealTime Systems // Proc. 15th Euromicro Conf. on RealTime Systems. Porto, Portugal, 2003.13. Ň·¯Ó‚ Ç.Ç. îÓÏËÓ‚‡ÌË ÂÍÓÏẨ‡ˆËÈ ÔÓ ËÁÏÂÌÂÌ˲ Ú·ӂ‡ÌËÈ Í ËÌÙÓχˆËÓÌÌÓÏÛ Ó·ÏÂÌÛÔË Ì‚ÓÁÏÓÊÌÓÒÚË ÔÓÒÚÓÂÌËfl ˆËÍÎÓ„‡ÏÏ˚ Ó·ÏÂÌÓ‚ ÔÓ ÏÛθÚËÔÎÂÍÒÌÓÏÛ Í‡Ì‡ÎÛ // í. íÂÚ¸ÂÈÏÂʉÛ̇.

ÍÓÌÙ. “è‡‡ÎÎÂθÌ˚ ‚˚˜ËÒÎÂÌËfl Ë Á‡‰‡˜Ë ÛÔ‡‚ÎÂÌËfl” (êÄëé'2006). å.: àèì ËÏ.Ç.Ä.í‡ÔÂÁÌËÍÓ‚‡ êÄç. 2006. ë. 422–437.14. ä‡Î‡¯ÌËÍÓ‚ Ä.Ç., äÓÒÚÂÌÍÓ Ç.Ä., å‡ÍËÌ å.à.ë‰ÒÚ‚‡ ÍÓÌÒÚÛËÓ‚‡ÌËfl ËÚÂ‡ˆËÓÌÌ˚ı ‡Î„ÓËÚÏÓ‚ ‰Îfl ¯ÂÌËfl Á‡‰‡˜ ÍÓÏ·Ë̇ÚÓÌÓÈ ÓÔÚËÏËÁ‡ˆËË // àÒÍÛÒÒÚ‚ÂÌÌ˚È ËÌÚÂÎÎÂÍÚ. 2004. ‹ 2.ë. 91–95.àáÇÖëíàü êÄç. íÖéêàü à ëàëíÖåõ ìèêÄÇãÖçàü      ‹ 6      2007.