Hardware-Transactional-Memory Based Speculative Parallel Discrete Event Simulation of Very Fine Grain Models
In previous papers, we have described a reduction model for computing near-perject state information (NPSI) in support of adaptive synchronization in a parallel discreteevent simulation. Here, we report on an implementation of this model on u popular high pe@ormmnce computing platfcnrm a network of workstutions without the use of special purpose hardware. The specific platform is u set of Pent&m Pro PC’s inter-connected by Myrinet a Gbps network. We describe the reduction model and its use in our Elastic Time Algorithm. We summarize our design. described in cm earlier paper and focus on the details of the implementation of this design. We present performance results that indicate that NPSI is feasible for simulations with medium to large event granular@.
Unfortunately, ACM prohibits us from displaying non-influential references for this paper.
To see the full reference list, please visit http://dl.acm.org/citation.cfm?id=278023.