ORCHESTRA: An asynchronous wait-free distributed GVT algorithm
- 1 October 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
Abstract
Taking advantage of computing capabilities offered by modern parallel and distributed architectures is fundamental to run large-scale simulation models based on the Parallel Discrete Event Simulation (PDES) paradigm. By relying on this computing organization, it is possible to effectively overcome both the power and the memory wall, which are core limiting aspects to deliver high-performance simulations. This is even more the case when relying on the speculative Time Warp synchronization protocol, which could be particularly memory greedy. At the same time, some form of coordination, such as the computation of the Global Virtual Time (GVT), is required by Time Warp Systems. These coordination points could easily become the bottleneck of large-scale simulations, hindering an efficient exploitation of the computing power offered by large supercomputing facilities. In this paper we present ORCHESTRA, a coordination algorithm which is both wait-free and asynchronous. The nature of this algorithm allows any computing node to carry on simulation activities while the global agreement is reached, thus offering an effective building block to achieve scalable PDES. We claim that the general organization of ORCHESTRA could be adopted by different high-performance computing applications, thus paving the way to a more effective usage of modern computing infrastructures.Keywords
This publication has 14 references indexed in Scilit:
- A Scalable GVT Estimation Algorithm for PDES: Using Lower Bound of Event-Bulk-TimeMathematical Problems in Engineering, 2015
- Consistent and efficient output-streams management in optimistic simulation platforms2013
- A Lightweight Heuristic-based Mechanism for Collecting Committed Consistent Global States in Optimistic Simulation2007
- Reflections on the memory wall2004
- Computing global virtual time in shared-memory multiprocessorsACM Transactions on Modeling and Computer Simulation, 1997
- Adaptive memory management and optimism control in time warpACM Transactions on Modeling and Computer Simulation, 1997
- Efficient Algorithms for Distributed Snapshots and Global Virtual Time ApproximationJournal of Parallel and Distributed Computing, 1993
- Wait-free synchronizationACM Transactions on Programming Languages and Systems, 1991
- On distributed snapshotsInformation Processing Letters, 1987
- Virtual timeACM Transactions on Programming Languages and Systems, 1985