Compiler-Based Data Prefetching and Streaming Non-temporal Store Generation for the Intel(R) Xeon Phi(TM) Coprocessor
Authors:
- Rakesh KrishnaiyerRKRakesh Krishnaiyer
- Emre KultursayEKEmre Kultursay
- Pankaj ChawlaPCPankaj Chawla
- Serguei PreisSPSerguei Preis
- Anatoly ZvezdinAZAnatoly Zvezdin
- Hideki SaitoHSHideki Saito
- 1 May 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1575-1586
Abstract
The Intel® Xeon Phi™ coprocessor has software prefetching instructions to hide memory latencies and special store instructions to save bandwidth on streaming non-temporal store operations. In this work, we provide details on compiler-based generation of these instructions and evaluate their impact on the performance of the Intel® Xeon Phi™ coprocessor using a wide range of parallel applications with different characteristics. Our results show that the Intel® Composer XE 2013 compiler can make effective use of these mechanisms to achieve significant performance improvements.Keywords
This publication has 18 references indexed in Scilit:
- Making data prefetch smarter2012
- A Case Study on Compiler Optimizations for the Intel® CoreTM 2 Duo ProcessorInternational Journal of Parallel Programming, 2008
- Dynamic speculative precomputation2005
- Effect of Optimizations on Performance of OpenMP ProgramsLecture Notes in Computer Science, 2004
- Designing a modern memory hierarchy with hardware prefetchingIEEE Transactions on Computers, 2001
- Data prefetch mechanismsACM Computing Surveys, 2000
- Compiler-based prefetching for recursive data structures1996
- An effective on-chip preloading scheme to reduce data access penalty1991
- Software prefetching1991
- Improving direct-mapped cache performance by the addition of a small fully-associative cache and prefetch buffers1990