ACM Transactions on Programming Languages and Systems (TOPLAS)
Parallel discrete event simulation
Communications of the ACM - Special issue on simulation
Adaptive checkpointing in Time Warp
PADS '94 Proceedings of the eighth workshop on Parallel and distributed simulation
Effects of the checkpoint interval on time and space in time warp
ACM Transactions on Modeling and Computer Simulation (TOMACS)
Comparative analysis of periodic state saving techniques in time warp simulators
PADS '95 Proceedings of the ninth workshop on Parallel and distributed simulation
A case study in simulating PCS networks using Time Warp
PADS '95 Proceedings of the ninth workshop on Parallel and distributed simulation
High performance messaging on workstations: Illinois fast messages (FM) for Myrinet
Supercomputing '95 Proceedings of the 1995 ACM/IEEE conference on Supercomputing
Exploiting model independence for parallel PCS network simulation
PADS '99 Proceedings of the thirteenth workshop on Parallel and distributed simulation
ROSS: a high-performance, low memory, modular time warp system
PADS '00 Proceedings of the fourteenth workshop on Parallel and distributed simulation
Proceedings of the 33nd conference on Winter simulation
MASCOTS '01 Proceedings of the Ninth International Symposium in Modeling, Analysis and Simulation of Computer and Telecommunication Systems
Nonblocking Checkpointing for Optimistic Parallel Simulation: Description and an Implementation
IEEE Transactions on Parallel and Distributed Systems
IEEE Transactions on Wireless Communications
Space uncertain simulation events: some concepts and an application to optimistic synchronization
Proceedings of the eighteenth workshop on Parallel and distributed simulation
Hi-index | 0.00 |
CCL (Checkpointing and Communication Library) is arecently developed software in support of optimistic parallelsimulation on myrinet based clusters. Beyond classicallow latency message delivery functionalities, this libraryimplements CPU offloaded, semi-asynchronous checkpointingfunctionalities based on data transfer capabilities providedby a programmable DMA engine on board of myrinetnetwork cards. The latest version of CCL (v2.4), designedfor M2M-PCI32C myrinet cards, only supports monoprogrammedsemi-asynchronous checkpoints. This forces resynchronizationbetween CPU and DMA activities each time a new checkpoint request must be issued at the simulation application level while the last issued one is still being carried out by the DMA engine. In this paper we present CCL v3.0 that, exploiting hardware features of more advanced M3M-PCI64C myrinet cards, supports multiprogrammed semi-asynchronous checkpoints. The multiprogrammed approach allows higher degree of concurrencybetween checkpointing and other simulation specific operations carried out by the CPU, with obvious benefits onperformance. We also report the results of the evaluationof those benefits for the case of a personal communicationsystem simulation application.