Distributed Event-driven Simulation- Scheduling Strategies and Resource Management Overeinder, B.J.

Transcription

1 UvA-DARE (Digital Academic Repository) Distributed Event-driven Simulation- Scheduling Strategies and Resource Management Overeinder, B.J. Link to publication Citation for published version (APA): Overeinder, B. J. (2000). Distributed Event-driven Simulation- Scheduling Strategies and Resource Management Amsterdam: University of Amsterdam General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. UvA-DARE is a service provided by the library of the University of Amsterdam ( Download date: 19 Jan 2019

2 Index x absolutee efficiency, 126 ACA,, see asynchronous cellular automata a activity,, 3 activityy scanning, 9 adaptivee optimism control globall state, 40 locall state, 39 aggressivee cancellation, 32 anti-message,, 31 APSE,, 84-91, 119 APSIS,, 49, 55-65, 114, 138 asynchronouss cellular automata, 105, nondeterministic,, 106 attribute,, 3 averagee parallelism, 77, 79, 91, 119 Beowulff parallel computer, 156 Boltzmannn distribution, 112 CA,, see cellular automata cancell queue, 66, 68 cancellationn strategies, 32 aggressive,, 32 direct,, 33 lazy,, 32 lazyy re-evaluation, 33 cascadedd rollbacks, 37, 130, 141 causality y (local)) constraint, 23, 61 error,, 22 cellularr automata, 104 asynchronous,, 105, 106 nondeterministic,, 106 parallell asynchronous, 108 parallell synchronous, 107 synchronous,, 104 checkpointing,, 168 clusterr computing, 156 clusterss of workstations, 156 computationall critical behavior, 137, computerr experiment, 1 computerr simulation, 1 conservativee methods, 20, continuouss system, 3 continuous-timee model, 6 copyy state saving, 34, 68 correlations,, 133, 137 long-range,, 137, 139, 142 COW,, see clusters of workstations criticall exponent, 133 criticall path, 78 analysis,, 83 enumerationn algorithm, 89 criticall sections, 168 criticall system, 135 criticall time, 83, 88 DASS parallel computer, 92,118,138, dataa dependency graph, 78 DCS,, see dynamic complex systems deadlockk avoidance, 25 deadlockk detection and recovery, 26 directt cancellation, 33 discretee event model, 7 discretee event simulation activityy scanning, 9 eventt scheduling, 8, 56 parallel,, 24, 28 processs interaction, 9, 56, 81 self-initiating,, 45 worldd view, 8 discretee system, 3 discrete-timee model, 7 DISS,, 98

3 220 0 Index Index distributedd computing, 11, 155 distributedd memory, 10, 156 DPVM,, see Dynamite dynamicc complex system, 103 dynamicc load balancing, see load balancing g Dynamicc PVM, see Dynamite Dynamite,, entity,, 3 event,, 3 endogenous,, 3 exogenous,, 3 external,, 21 internal,, 21 scheduling,, 8 eventt granularity, 127 eventt message, 21 eventt precedence graph, 80 eventt queue, see input queue eventt retraction, 66, 71 event-drivenn simulation, 8 experimentall framework, 3 Fornax,, 52 fossill collection, 31, 41, 72 fractall properties, 133 globall virtual time, 31, 41, 61, 73 centralizedd computation, 41 distributedd computation, 43 gridd computing, 154 GVT,, see global virtual time Highh Level Architecture, 13 highh performance computing, 12,158 highh throughput computing, 12, 158 HLA,, see High Level Architecture incrementall state saving, 35, 67 inherentt parallelism, 77, 119 inputt queue, 63, 68 Isingg spin model, 109, 135 continuous-time,, 113, 114 criticall temperature, 113 ferromagnetic,, 110 paramagnetic,, 110 Isingg spin phase transition, 137, 138 job,, 11 lazyy cancellation, 32 lazyy re-evaluation, 33 loadd balancing, 12 dynamic,, 12, 162 local-area,, 166 migrationn decider, 167 optimisticc simulation, 39, 190 resourcee monitoring, 167 scheduler,, 166 wide-area,, 161 locall virtual time, 30, 61 local-areaa networks, 156 logicall clocks, 30 logicall processes, 21 long-rangee correlations, 137, 139, lookahead,, 27 LVT,, see local virtual time massivelyy parallel processor, 10,156 metacomputingg environment, 154 Metropoliss algorithm, 112 model,, 4 continuous,, 5 deterministic,, 5 discrete,, 5 dynamic,, 5 state,, 3, 21 static,, 5 stochastic,, 5 Montee Carlo method, 103, 112, 126 Montee Carlo time step, 119 MPP,, see massive parallel processor networkss of workstations, 156 non-adaptivee optimism control, 38 NOW,, see networks of workstations nulll message, 25 optimismm control, 37, 62, 70, 124, adaptivee global state, 40 adaptivee local state, 39 non-adaptive,, 38

4 Index Index optimismm throttling, see optimism control,, 124 optimisticc methods, 20, outputt queue, 63, 68 parallell discrete event simulation, 20 0 conservative,, 24 optimistic,, 28 parallell efficiency absolute,, 126 relative,, 118 parallell random number generation, leap-frog,, 116 splitting,, 116 parallelismm profile, 79, 90 periodicc checkpointing, see periodic statee saving periodicc state saving, 34 PERTT algorithm, 87, 88 physicall critical behavior, 137, 150 physicall processes, 21 Polderr metacomputer, positivee message, 31 potentiall parallelism, 77 power-laww distribution, 133 predictability,, 26 priorityy queue, 69 process,, 11 processs interaction, 9 processs migration, 165, 169 programm activity graph, 83, 88 pseudo-randomm number generator, PVM,, 164 daemons,, 164 tasks,, 164 randomm number generator, 115 relativee efficiency, 118 resourcee management, 11, 154, 160 rollbackk strategies, see cancellation strategies s scale-invariantt properties, 133 scalingg exponent, 133 scheduling,, 12 dynamic,, 12, 162 global,, 12, 161 local,, 12 optimisticc simulation, 39 static,, 12, 162 self-organizedd criticality, 133 finite-sizefinite-size scaling, 142 self-similarr properties, 133 seperationn of time scales, 135 sequentiall fraction, 78, 91 shapee vector, 79, 90 sharedd memory, 10 simulation n continuous,, 5 discretee event, 5 event-driven,, 8 Montee Carlo, 5 predictability,, 4 realizability,, 4 time-driven,, 7 welll defined, 4 simulationn languages, 51 simulationn libraries, 53 simulationn model, see model simultaneouss events, 70 SOC,, see self-organized criticality space-timee diagram, 81 speedupp models absolute,, 126 memory-bounded,, 67, 78 relative,, 118 statee queue, 63, 68 statee saving, 34 copy,, 34, 68 hybrid,, 37 incremental,, 35, 67 periodic,, 34 straggler,, 30 super-criticall speedup, 33, 99 system,, 3 componentss of a, 3 continuous,, 3 discrete,3 3 environment,, 3 state,, 3

5 222 2 Index Index task,, 11 taskk precedence graph, 78 thrashing,, 35, 37, 123 throttling,, 35, see optimism control Timee Warp, 28, 30-31, 56, see AP- SIS S time-drivenn simulation, 7 timestamp,, 21 validation,, 2 virtuall environments, 13, 187 virtuall time, virtuall time window, 124, 147, see optimismm control well-defined,, 26 wide-areaa networks, 157 worldd view, 8 activityy scanning, 9 eventt scheduling, 8, 56 processs interaction, 9, 56, 81

6

7