Change search
ReferencesLink to record
Permanent link

Direct link
MANY-TASK COMPUTING ON MANY-CORE ARCHITECTURES
KTH, School of Computer Science and Communication (CSC), Computational Science and Technology (CST). BCAM, Bilbao, Spain.
Show others and affiliations
2016 (English)In: Scalable Computing: Practice and Experience, ISSN 1895-1767, Vol. 17, no 1, 33-46 p.Article in journal (Refereed) PublishedText
Abstract [en]

Many-Task Computing (MTC) is a common scenario for multiple parallel systems, such as cluster, grids, cloud and supercomputers, but it is not so popular in shared memory parallel processors. In this sense and given the spectacular growth in performance and in number of cores integrated in many-core architectures, the study of MTC on such architectures is becoming more and more relevant. In this paper, authors present what are those programming mechanisms to take advantages of such massively parallel features for the particular target of MTC. Also, the hardware features of the two dominant many-core platforms (NVIDIA's GPUs and Intel Xeon Phi) are also analyzed for our specific framework. Given the important differences in terms of hardware and software in our two many-core platforms, we have considered different strategies based on CUDA (for GPUs) and OpenMP (for Intel Xeon Phi). We carried out several test cases based on an appropriate and widely studied problem for benchmarking as matrix multiplication. Essentially, this study consisted of comparing the time consumed for computing in parallel several tasks one by one (the whole computational resources are used just to compute one task at a time) with the time consumed for computing in parallel the same set of tasks simultaneously (the whole computational resources are used for computing the set of tasks at very same time). Finally, we compared both software-hardware scenarios to identify the most relevant computer features in each of our many-core architectures.

Place, publisher, year, edition, pages
UNIV VEST TIMISOARA , 2016. Vol. 17, no 1, 33-46 p.
Keyword [en]
Parallel Computing, Multi-Task Computing, Many-Core, GPU, Intel Xeon Phi, CUDA, OpenMP
National Category
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-185655ISI: 000373067600004ScopusID: 2-s2.0-84963752839OAI: oai:DiVA.org:kth-185655DiVA: diva2:923525
Note

QC 20160426

Available from: 2016-04-26 Created: 2016-04-25 Last updated: 2016-04-26Bibliographically approved

Open Access in DiVA

No full text

Scopus

Search in DiVA

By author/editor
Jansson, Johan
By organisation
Computational Science and Technology (CST)
In the same journal
Scalable Computing: Practice and Experience
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 12 hits
ReferencesLink to record
Permanent link

Direct link