%0 Conference Proceedings
%T MAP-numa: Access Patterns Used to Characterize the NUMA Memory Access Optimization Techniques and Algorithms
%+ National High Performance Computing Center (NHPCC)
%+ Shenzhen University [Shenzhen]
%+ Chinese Academy of Sciences [Beijing] (CAS)
%A Luo, Qiuming
%A Liu, Chenjian
%A Kong, Chang
%A Cai, Ye
%Z Part 5: Performance Modeling, Prediction, and Tuning
%< avec comité de lecture
%( Lecture Notes in Computer Science
%B 9th International Conference on Network and Parallel Computing (NPC)
%C Gwangju, South Korea
%Y James J. Park
%Y Albert Zomaya
%Y Sang-Soo Yeo
%Y Sartaj Sahni
%I Springer
%3 Network and Parallel Computing
%V LNCS-7513
%P 208-216
%8 2012-09-06
%D 2012
%R 10.1007/978-3-642-35606-3_24
%K NUMA
%K Benchmark
%K Oprofile
%K Memory Access Patterns
%Z Computer Science [cs]Conference papers
%X Some typical memory access patterns are provided and programmed in C, which can be used as benchmark to characterize the various techniques and algorithms aim to improve the performance of NUMA memory access. These access patterns, called MAP-numa (Memory Access Patterns for NUMA), currently include three classes, whose working data sets are corresponding to 1-dimension array, 2-dimension matrix and 3-dimension cube. It is dedicated for NUMA memory access optimization other than measuring the memory bandwidth and latency. MAP-numa is an alternative to those exist benchmarks such as STREAM, pChase, etc. It is used to verify the optimizations’ (made automatically/manually to source code/executive binary) capacities by investigating what locality leakage can be remedied. Some experiment results are shown, which give an example of using MAP-numa to evaluate some optimizations based on Oprofile sampling.
%G English
%Z TC 10
%Z WG 10.3
%2 https://inria.hal.science/hal-01551378/document
%2 https://inria.hal.science/hal-01551378/file/978-3-642-35606-3_24_Chapter.pdf
%L hal-01551378
%U https://inria.hal.science/hal-01551378
%~ IFIP-LNCS
%~ IFIP
%~ IFIP-AICT
%~ IFIP-TC
%~ IFIP-TC10
%~ IFIP-NPC
%~ IFIP-WG10-3
%~ IFIP-LNCS-7513