AdaptMemBench: Application-Specific MemorySubsystem Benchmarking

Mahesh Lakshminarasimhan; Catherine Olschanowsky

AdaptMemBench: Application-Specific MemorySubsystem Benchmarking

Performance 2018-12-20 v1

Authors: Mahesh Lakshminarasimhan , Catherine Olschanowsky

Abstract

Optimizing scientific applications to take full advan-tage of modern memory subsystems is a continual challenge forapplication and compiler developers. Factors beyond working setsize affect performance. A benchmark framework that exploresthe performance in an application-specific manner is essential tocharacterize memory performance and at the same time informmemory-efficient coding practices. We present AdaptMemBench,a configurable benchmark framework that measures achievedmemory performance by emulating application-specific accesspatterns with a set of kernel-independent driver templates. Thisframework can explore the performance characteristics of a widerange of access patterns and can be used as a testbed for potentialoptimizations due to the flexibility of polyhedral code generation.We demonstrate the effectiveness of AdaptMemBench with casestudies on commonly used computational kernels such as triadand multidimensional stencil patterns.

Keywords

computer architecture memory hierarchy processing-in-memory

Cite

@article{arxiv.1812.07778,
  title  = {AdaptMemBench: Application-Specific MemorySubsystem Benchmarking},
  author = {Mahesh Lakshminarasimhan and Catherine Olschanowsky},
  journal= {arXiv preprint arXiv:1812.07778},
  year   = {2018}
}

AdaptMemBench: Application-Specific MemorySubsystem Benchmarking

Abstract

Keywords

Cite

Related papers