Heterogeneous Voltage Frequency Scaling of Data-Parallel Applications for Energy Saving on Homogeneous Multicore Platforms

被引：4

作者：

Bratek, Pawel ^{[1
]}

Szustak, Lukasz ^{[1
]}

Wyrzykowski, Roman ^{[1
]}

Olas, Tomasz ^{[1
]}

Chmiel, Tomasz ^{[1
]}

机构：

[1] Czestochowa Tech Univ, Dept Comp Sci, Dabrowskiego 69, PL-42201 Czestochowa, Poland

来源：

EURO-PAR 2021: PARALLEL PROCESSING WORKSHOPS | 2022年 / 13098卷

关键词：

Data-parallel applications; Energy saving; Heterogeneous voltage frequency scaling; Multicore; ccNUMA; PERFORMANCE;

D O I：

10.1007/978-3-031-06156-1_12

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

In this paper, for the first time, we explore and establish the combined benefits of heterogeneous DVFS (dynamic voltage frequency scaling) control in improving the energy-performance behavior of data-parallel applications on shared-memory multicore systems. We propose to customize the clock frequency individually for the appropriately selected groups of cores corresponding to the diversified time of actual computation. In consequence, the advantage of up to 20% points over the homogeneous frequency scaling is achieved on the ccNUMA server with two 18-core Intel Xeon Gold 6240 containing 72 logical cores in total. The cost and efficiency of the proposed pruning algorithm for selecting heterogeneous DVFS configurations against the brute-force search are verified and compared experimentally.

引用

页码：141 / 153

页数：13

共 20 条

[1] Performance-Energy Trade-off in CMPs with Per-Core DVFS [J].

Abera, Solomon ;

Balakrishnan, M. ;

Kumar, Anshul .

ARCHITECTURE OF COMPUTING SYSTEMS, 2018, 10793 :225-238

[2]

Acun B., 2019, 2019 10 INT GREEN SU, V1, P1

[3]

[Anonymous], 2021, TECHNOLOGY GUIDE INT

[4]

Calore Enrico, 2018, Journal of Low Power Electronics and Applications, V8, DOI 10.3390/jlpea8020018

[5]

Ciznicki M., 2015, HISTENCIL 2015 2 INT, P943

[6]

Crank J., 1979, MATH DIFFUSION

[7]

Gupta M., 2020, P 4 INT C IOT SOC MO, P1201

[8]

HajYahya J, 2018, COMPUT ARCHIT DES ME, P1, DOI 10.1007/978-981-10-8554-3

[9]

Kolpe T, 2011, DES AUT TEST EUROPE, P293

[10] Quantifying the energy efficiency challenges of achieving exascale computing [J].

Mair, Jason ;

Huang, Zhiyi ;

Eyers, David ;

Chen, Yawen .

2015 15TH IEEE/ACM INTERNATIONAL SYMPOSIUM ON CLUSTER, CLOUD AND GRID COMPUTING, 2015, :943-950

← 1 2 →