Autotuning Runtime Specialization for Sparse Matrix-Vector Multiplication

被引：10

作者：

Yilmaz, Buse ^{[1
]}

Aktemur, Baris ^{[1
]}

Garzaran, Maria J. ^{[2
,3
,5
]}

Kamin, Sam ^{[2
,4
,5
]}

Kirac, Furkan ^{[1
]}

机构：

[1] Ozyegin Univ, TR-34794 Istanbul, Turkey

[2] Univ Illinois, Urbana, IL USA

[3] Intel Corp, Santa Clara, CA 95051 USA

[4] Google, Bangalore, Karnataka, India

[5] Univ Illinois, 201 N Goodwin Ave, Urbana, IL 61801 USA

来源：

ACM TRANSACTIONS ON ARCHITECTURE AND CODE OPTIMIZATION | 2016年 / 13卷 / 01期

基金：

美国国家科学基金会;

关键词：

Performance; Experimentation; Measurement; Autotuning; runtime code generation; sparse matrix-vector multiplication; GENERATION;

D O I：

10.1145/2851500

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Runtime specialization is used for optimizing programs based on partial information available only at runtime. In this paper we apply autotuning on runtime specialization of Sparse Matrix-Vector Multiplication to predict a best specialization method among several. In 91% to 96% of the predictions, either the best or the second-best method is chosen. Predictions achieve average speedups that are very close to the speedups achievable when only the best methods are used. By using an efficient code generator and a carefully designed set of matrix features, we show the runtime costs can be amortized to bring performance benefits for many real-world cases.

引用

页数：26

共 50 条

[1] Optimization by Runtime Specialization for Sparse Matrix-Vector Multiplication
Kamin, Sam
Garzaran, Maria Jesus
Aktemur, Baris
Xu, Danqing
Yilmaz, Buse
Chen, Zhongbo
[J]. ACM SIGPLAN NOTICES, 2015, 50 (03) : 93 - 102
[2] Sparse Matrix-Vector Multiplication on GPGPUs
Filippone, Salvatore
Cardellini, Valeria
Barbieri, Davide
Fanfarillo, Alessandro
[J]. ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 2017, 43 (04):
[3] GPU accelerated sparse matrix-vector multiplication and sparse matrix-transpose vector multiplication
Tao, Yuan
Deng, Yangdong
Mu, Shuai
Zhang, Zhenzhong
Zhu, Mingfa
Xiao, Limin
Ruan, Li
[J]. CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2015, 27 (14): : 3771 - 3789
[4] Adaptive Runtime Tuning of Parallel Sparse Matrix-Vector Multiplication on Distributed Memory Systems
Lee, Seyong
Eigenmann, Rudolf
[J]. ICS'08: PROCEEDINGS OF THE 2008 ACM INTERNATIONAL CONFERENCE ON SUPERCOMPUTING, 2008, : 195 - 204
[5] Vector ISA extension for sparse matrix-vector multiplication
Vassiliadis, S
Cotofana, S
Stathis, P
[J]. EURO-PAR'99: PARALLEL PROCESSING, 1999, 1685 : 708 - 715
[6] Node aware sparse matrix-vector multiplication
Bienz, Amanda
Gropp, William D.
Olson, Luke N.
[J]. JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2019, 130 : 166 - 178
[7] Sparse Matrix-Vector Multiplication on a Reconfigurable Supercomputer
DuBois, David
DuBois, Andrew
Connor, Carolyn
Poole, Steve
[J]. PROCEEDINGS OF THE SIXTEENTH IEEE SYMPOSIUM ON FIELD-PROGRAMMABLE CUSTOM COMPUTING MACHINES, 2008, : 239 - +
[8] Understanding the performance of sparse matrix-vector multiplication
Goumas, Georgios
Kourtis, Kornilios
Anastopoulos, Nikos
Karakasis, Vasileios
Koziris, Nectarios
[J]. PROCEEDINGS OF THE 16TH EUROMICRO CONFERENCE ON PARALLEL, DISTRIBUTED AND NETWORK-BASED PROCESSING, 2008, : 283 - +
[9] Sparse matrix-vector multiplication design on FPGAs
Sun, Junqing
Peterson, Gregory
Storaasli, Olaf
[J]. FCCM 2007: 15TH ANNUAL IEEE SYMPOSIUM ON FIELD-PROGRAMMABLE CUSTOM COMPUTING MACHINES, PROCEEDINGS, 2007, : 349 - +
[10] STRUCTURED SPARSE MATRIX-VECTOR MULTIPLICATION ON A MASPAR
DEHN, T
EIERMANN, M
GIEBERMANN, K
SPERLING, V
[J]. ZEITSCHRIFT FUR ANGEWANDTE MATHEMATIK UND MECHANIK, 1994, 74 (06): : T534 - T538

← 1 2 3 4 5 →