Performance Characteristics of Hardware Transactional Memory for Molecular Dynamics Application on BlueGene/Q: Toward Efficient Multithreading Strategies for Large-Scale Scientific Applications
Authors:
- Manaschai KunasethMKManaschai Kunaseth
- Rajiv K. KaliaRKRajiv K. Kalia
- Aiichiro NakanoANAiichiro Nakano
- Priya VashishtaPVPriya Vashishta
- David F. RichardsDRDavid F. Richards
- James N. GlosliJGJames N. Glosli
- 1 May 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1326-1335
Abstract
We have investigated the performance characteristics of hardware transactional memory (HTM) on the Blue Gene/Q computer in comparison with conventional concurrency control mechanisms, using a molecular dynamics application as an example. Benchmark tests, along with overhead-cost and scalability analysis, quantify relative performance advantages of HTM over other mechanisms. We found that the bookkeeping cost of HTM is high but that the rollback cost is low. We propose transaction fusion and spatially-compact scheduling techniques to reduce the overhead of HTM with minimal programming. A strong scalability benchmark shows that the fused HTM has the shortest runtime among various concurrency control mechanisms without extra memory. Based on the performance characterization, we derive a decision tree in the concurrency-control design space for multithreading application.Keywords
This publication has 15 references indexed in Scilit:
- The IBM Blue Gene/Q Compute ChipIEEE Micro, 2011
- Optimizing hybrid transactional memory2011
- Computing Performance: Game Over or Next Level?Computer, 2011
- RobuSTM: A Robust Software Transactional Memory2010
- Time-Based Software Transactional MemoryIEEE Transactions on Parallel and Distributed Systems, 2010
- Millisecond-scale molecular dynamics simulations on Anton2009
- Rock: A High-Performance Sparc CMT ProcessorIEEE Micro, 2009
- Extending stability beyond CPU millennium2007
- The fast Fourier Poisson method for calculating Ewald sumsThe Journal of Chemical Physics, 1994
- Particle mesh Ewald: An N⋅log(N) method for Ewald sums in large systemsThe Journal of Chemical Physics, 1993