Towards Automated Memory Model Generation Via Event Tracing

Journal Article · Mon Jun 04 04:00:00 EDT 2012 · Computer Journal

DOI:https://doi.org/10.1093/comjnl/bxs051· OSTI ID:1078671

Perks, O. F. J.; Beckingsale, D. A.; Hammond, S. D.; Miller, I.; Herdman, J. A.; Vadgama, A.; Bhalerao, A. H.; He, L.; Jarvis, S. A.

The importance of memory performance and capacity is a growing concern for high performance computing laboratories around the world. It has long been recognized that improvements in processor speed exceed the rate of improvement in dynamic random access memory speed and, as a result, memory access times can be the limiting factor in high performance scientific codes. The use of multi-core processors exacerbates this problem with the rapid growth in the number of cores not being matched by similar improvements in memory capacity, increasing the likelihood of memory contention. In this paper, we present WMTools , a lightweight memory tracing tool and analysis framework for parallel codes, which is able to identify peak memory usage and also analyse per-function memory use over time. An evaluation of WMTools , in terms of its effectiveness and also its overheads, is performed using nine established scientific applications/benchmark codes representing a variety of programming languages and scientific domains. We also show how WMTools can be used to automatically generate a parameterized memory model for one of these applications, a two-dimensional non-linear magnetohydrodynamics application, Lare2D . Through the memory model we are able to identify an unexpected growth term which becomes dominant at scale. With a refined model we are able to predict memory consumption with under 7% error.

• Research Organization: Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
• Sponsoring Organization: USDOE National Nuclear Security Administration (NNSA)
• DOE Contract Number: AC04-94AL85000
• OSTI ID: 1078671
• Report Number(s): SAND2012--0920J; 447937
• Journal Information: Computer Journal, Journal Name: Computer Journal Journal Issue: 2 Vol. 56; ISSN 0010-4620
• Country of Publication: United States
• Language: English

Related Subjects

97 MATHEMATICS AND COMPUTING
high performance computing
memory
modeling
multi-core
tracing