Virtual Time III, Part 3: Throttling and Message Cancellation

被引:0
作者
Jefferson, David [1 ]
Barnes, Peter d. [2 ]
机构
[1] Lawrence Livermore Natl Lab, Ctr Appl Sci Comp, Livermore, CA 94550 USA
[2] Lawrence Livermore Natl Lab, Livermore, CA USA
来源
ACM TRANSACTIONS ON MODELING AND COMPUTER SIMULATION | 2024年 / 34卷 / 04期
关键词
Parallel discrete event simulation; virtual time; synchronization; conser-; vative; optimistic; throttling; invariant; monotonicity; rollback; antimessage; lookahead; Unified Virtual Time; UVT;
D O I
10.1145/3678173
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
This is Part 3 of a trio of papers that unify in a natural way the two historically distinct parallel discrete event synchronization paradigms, optimistic and conservative, combining the best properties of both into a single framework called Unified Virtual Time (U V T) . In this part, we survey the synchronization effects that can be achieved by restricting to corner cases the relationships permitted among the control variables, GVT , CVT , TVT , and [VT , which were defined in Part 1 . We also survey various throttling policies from the literature and describe how they can be implemented in U V T by controlling the value of TVT , including policies that can take advantage of rollback in addition to LP blocking. A significant result is a new category of efficient and higher precision throttling algorithms for optimistic execution that are based on optimistic lookahead , defined in a way that is symmetric to what we now call the conservative lookahead information that is traditionally used for conservative synchronization. Finally, we present a novel algorithm allowing the choice between lazy and aggressive cancellation to be made on a message-by-message basis using either external logic expressed in the model code, or policy code internal to the simulator, or a mixture of both. CCS Concepts: center dot Computing methodologies - Modeling and simulation ; Simulation types and techniques ; Discrete-event simulation ; Distributed simulation ; Massively parallel and highperformance simulations ; Distributed computing methodologies; center dot Theory of computation - Design and analysis of algorithms ; Parallel algorithms ; Massively parallel algorithms ;
引用
收藏
页数:36
相关论文
共 46 条
[1]  
Abu-Ghazaleh NB, 2006, IEEE ACM DIS SIM, P253
[2]  
Ball D., 1990, Distributed Simulation. Proceedings of the SCS Multiconference, P174
[3]  
Berry O., 1985, P 1985 SCS C DISTR S
[4]  
Berry O., 1986, Ph.D. Dissertation
[5]   DISTRIBUTED SIMULATION - CASE-STUDY IN DESIGN AND VERIFICATION OF DISTRIBUTED PROGRAMS [J].
CHANDY, KM ;
MISRA, J .
IEEE TRANSACTIONS ON SOFTWARE ENGINEERING, 1979, 5 (05) :440-452
[6]  
Das S., 1997, ACM Transactions on Modeling and Computer Simulation (TOMACS), V7, P2
[7]  
Das SR, 1996, 1996 WINTER SIMULATION CONFERENCE PROCEEDINGS, P186, DOI 10.1145/256562.256602
[8]  
Das SR, 1996, TENTH WORKSHOP ON PARALLEL AND DISTRIBUTED SIMULATION - PADS 96, PROCEEDINGS, P186
[9]  
Ferscha A., 1995, Proceedings. Ninth Workshop on Parallel and Distributed Simulation. (PADS'95) (Cat. No.95TB8096), P120, DOI 10.1109/PADS.1995.404309
[10]  
Ferscha A., 1995, P 28 ANN SIM S SS 95