-
Comparative analysis of coprocessors
- Back
Document Title
Comparative analysis of coprocessors
Author
Sakdhnagool P, Sabne A, Eigenmann R
Name from Authors Collection
Affiliations
National Science & Technology Development Agency - Thailand; National Electronics & Computer Technology Center (NECTEC); Microsoft; University of Delaware
Type
Article
Source Title
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
ISSN
1532-0626
Year
2019
Volume
31
Issue
1
Page
-
Open Access
Bronze
Publisher
WILEY
DOI
10.1002/cpe.4756
Format
Abstract
While GPUs have seen a steady increase in usage, Xeon Phis have struggled in proving their value, and eventually got discontinued. Is this a matter of the Intel many-core architecture's younger age or are there reasons due to specific features? This paper reviews quantitative information addressing these questions. Using two latest coprocessors, we evaluate performance and programming productivity across a range of microbenchmarks and applications. We consider productivity as the percentage of hand-optimized performance reached by a simple high-level parallel programming model that is translated onto the specific architectures by advanced compilers. We evaluate and compare the performance of the two coprocessors and point out where Xeon Phis fall short. We also briefly review the performance of different execution modes of Xeon Phis. Moreover, unlike common expectation, we found that Xeon Phis' productivity is marginally better that GPUs. The current results suggest that the performance advantage of GPUs outweighs the productivity benefit of Xeon Phis. Closing the performance gap and increasing the productivity benefit that a more regular many-core paradigm can offer will be essential in designing a next-generation architecture.
Keyword
accelerator | GPU | Performance | productivity | Xeon Phi
Industrial Classification
Knowledge Taxonomy Level 1
Knowledge Taxonomy Level 2
Funding Sponsor
National Science Foundation
License
Copyright
Rights
Publisher
Publication Source
WOS