COTSon: infrastructure for full system simulation
ACM SIGOPS Operating Systems Review
SlackSim: a platform for parallel simulations of CMPs on CMPs
ACM SIGARCH Computer Architecture News
Instruction-level simulation of a cluster at scale
Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis
Software—Practice & Experience
An analysis of queuing network simulation using GPU-based hardware acceleration
ACM Transactions on Modeling and Computer Simulation (TOMACS)
Adaptive and Speculative Slack Simulations of CMPs on CMPs
MICRO '43 Proceedings of the 2010 43rd Annual IEEE/ACM International Symposium on Microarchitecture
VSim: Simulating multi-server setups at near native hardware speed
ACM Transactions on Architecture and Code Optimization (TACO) - HIPEAC Papers
A survey on cache tuning from a power/energy perspective
ACM Computing Surveys (CSUR)
Proceedings of the ACM International Conference on Computing Frontiers
Hi-index | 0.00 |
Computer clusters are a very cost-effective approach for high performance computing, but simulating a complete cluster is still an open research problem. The obvious approach - to parallelize individual node simulators - is complex and slow. Combining individual parallel simulators implies synchronizing their progress of time. This can be accomplished with a variety of parallel discrete event simulation techniques, but unfortunately any straightforward approach introduces a synchronization overhead causing up two orders of magnitude of slowdown with respect to the simulation speed of an individual node. In this paper we present a novel adaptive technique that automatically adjusts the synchronization boundaries. By dynamically relaxing accuracy over the least interesting computational phases we dramatically increase performance with a marginal loss of precision. For example, in the simulation of an 8-node cluster running NAMD (a parallel molecular dynamics application) we show an acceleration factor of 26x over the deterministic "ground truth" simulation, at less than a 1% accuracy error.