Evaluation of HCM: A New Model to Predict the Execution Time of Regular Parallel Applications on a Heterogeneous Cluster

被引：0

作者：

Soares, Thiago Marques ^{[1
]}

dos Santos, Rodrigo Weber ^{[1
]}

Lobosco, Marcelo ^{[1
]}

机构：

[1] Univ Fed Juiz de Fora, Grad Program Computat Modeling, Juiz De Fora, Brazil

来源：

PARALLEL PROCESSING AND APPLIED MATHEMATICS (PPAM 2017), PT II | 2018年 / 10778卷

关键词：

Performance modeling; Parallel architectures; Heterogeneous clusters; Scheduling; Resource management;

D O I：

10.1007/978-3-319-78054-2_6

中图分类号：

TP31 [计算机软件];

学科分类号：

081202 ; 0835 ;

摘要：

In a previous work we proposed a new model that predicts the execution time of a regular application on a heterogeneous parallel environment. The model considers that a heterogeneous cluster is composed by distinct types of processors, accelerators and networks. This work further details and proposes some modifications to the original model, as well as evaluate it on a heterogeneous cluster environment. The results have show that the worst error in the estimations of the parallel execution time was about 12.7%, and, in many cases, the estimated execution time is equal to or very close to the actual one.

引用

页码：58 / 67

页数：10

共 50 条

[31] Matching and scheduling algorithms for minimizing execution time and failure probability of applications in heterogeneous computing
Dogan, A
Özgüner, F
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2002, 13 (03) : 308 - 323
[32] SHARP: A new real-time scheduling algorithm to improve security of parallel applications on heterogeneous clusters
Xie, Tao
Qin, Xiao
Nijim, Mais
2006 IEEE INTERNATIONAL PERFORMANCE COMPUTING AND COMMUNICATIONS CONFERENCE, VOLS 1 AND 2, 2006, : 39 - +
[33] HLogGP: A new parallel computational model for heterogeneous clusters
Bosque, JL
Perez, LP
2004 IEEE INTERNATIONAL SYMPOSIUM ON CLUSTER COMPUTING AND THE GRID - CCGRID 2004, 2004, : 403 - 410
[34] A green SLA constrained scheduling algorithm for parallel/scientific applications in heterogeneous cluster systems
Kumar, Neetesh
Vidyarthi, Deo Prakash
SUSTAINABLE COMPUTING-INFORMATICS & SYSTEMS, 2019, 22 : 107 - 119
[35] Run-time support for scheduling parallel applications in heterogeneous NOWs
Weissman, JB
Zhao, X
SIXTH IEEE INTERNATIONAL SYMPOSIUM ON HIGH PERFORMANCE DISTRIBUTED COMPUTING, PROCEEDINGS, 1997, : 347 - 355
[36] A cost model for the estimation of query execution time in a parallel environment supporting pipeline
Spiliopoulou, M
Hatzopoulos, M
Vassilakis, C
COMPUTERS AND ARTIFICIAL INTELLIGENCE, 1996, 15 (04): : 341 - 368
[37] Multi-paradigm and multi-grain parallel execution model based on SMP-cluster
Wang, Jue
Hu, Changjun
Lai, Jianxin
Zhao, Yudi
Zhang, Suqin
IEEE JOHN VINCENT ATANASOFF 2006 INTERNATIONAL SYMPOSIUM ON MODERN COMPUTING, PROCEEDINGS, 2006, : 266 - +
[38] Design and evaluation of a new transaction execution model for multidatabase systems
Devirmis, T
Ulusoy, O
INFORMATION SCIENCES, 1997, 102 (1-4) : 203 - 238
[39] A new model for continuous evaluation of suppliers with real execution data
Tekez, Esra Kurt
Kar, Fatih
Toklu, Merve Cengiz
JOURNAL OF ENGINEERING RESEARCH, 2019, 7 (02): : 298 - 314
[40] Execution time predicition model for parallel GPU realization of dicscrete transforms computation algorithms
Puchala, Dariusz
Stokfiszewski, Kamil
Wieloch, Kamil
BULLETIN OF THE POLISH ACADEMY OF SCIENCES-TECHNICAL SCIENCES, 2022, 70 (01)

← 1 2 3 4 5 →