Accelerating architectural simulation by parallel execution of trace samples
Gary Lauterbach · 1994
In order to quickly decide which architectural features are to be included in future processors the author has developed a simulation approach that samples benchmark program instruction traces. Rather than simulating an architecture with the entire SPEC92 program suite of more than 100 billion instructions, he simulates using a set of samples of the SPEC92 suite containing less than 1% of the total instruction trace. Each of the samples contains a short instruction trace that can be simulated quickly. By distributing the simulation of the samples across many workstations he is able to carry out architectural simulations in less than one half hour. The sample set is verified to be representative of the complete instruction trace using several metrics. The technique described can be applied to existing architectural models to produce significant reductions in simulation time.>
Cited by 64 works. doi:10.1109/hicss.1994.323171
Authors
References
- The AWK Programming Language
- Shade: a fast instruction-set simulator for execution profiling
- Aspects of Cache Memory and Instruction Buffer Performance
- Improving the accuracy of dynamic branch prediction using branch correlation
- High-bandwidth data memory systems for superscalar processors
- Accurate low-cost methods for performance evaluation of cache memory systems
- The SPARC architecture manual (version 9)
- Shade: a fast instruction-set simulator for execution profiling
- A comparison of trace-sampling techniques for multi-megabyte caches
- SpixTools: Introduction and User's Manual