skip to main content
10.1145/1378533.1378567acmconferencesArticle/Chapter ViewAbstractPublication PagesspaaConference Proceedingsconference-collections
research-article

A temperature-aware virtual submesh allocation scheme for noc-based manycore chips

Published: 14 June 2008 Publication History

Abstract

Various continuous and non-continuous submesh allocation schemes have been proposed for traditional mesh-connected multiprocessor systems. Discussions of these schemes are limited to issues such as fragmentation, system performance and algorithmic complexity. NoC-based manycore chips with mesh topology are envisaged to enter the domain of parallel computing and cores of them are allocated to applications in the forms of submeshes. However, unlike the processors in traditional systems, these cores may face severe runtime thermal non-uniformities and a promising strategy to avoid these thermal crises is keeping good heat balance throughout chips at runtime. One opportunity for good heat balance is to include thermally favorable cores when submeshes are allocated. This requires a suitable temperature sensitive submesh allocation scheme for manycore chips.
We propose a novel temperature-aware virtual submesh allocation scheme for NoC-based manycore chips. This scheme considers transient temperatures of cores when OSes allocate submeshes. Further, this scheme accommodates jobs with dynamically constructed virtual submeshes that are different from physical submeshes in existing schemes. In a virtual submesh, cores belonging to a virtual column (row) may come from different physical columns (rows). When chosen into virtual submeshes, cores with lower temperatures are preferred to ones with higher temperatures. Thus, cores which are not occupied and not overheated are considered for allocation; overheated cores can cool down during an idle or sleeping period. Our scheme aims to accommodate a job with a virtual submesh with low average temperature, called a cool virtual submesh.
Our scheme constructs a cool virtual submesh in steps to reduce complexity since finding an optimal submesh allocation is NP-complete. Firstly, we adopt algorithm, namely GCR, to construct a suitable-sized virtual submesh for a job J. If submeshes by GCR can not satisfy the request of this job even after switching request orientation, our scheme keeps this job waiting. Secondly, we construct a cool virtual submesh based on the virtual submesh constructed by GCR if it satisfy the request of this job. We dynamically construct "areas" with left and right boundaries. A local optimal cool virtual column which is the virtual column having the lowest average temperature of cores is to be searched in each of these areas. We search local optimal cool virtual columns one by one from right to left. We model this searching problem as finding the shortest weighted paths of a directed network. We develop an algorithm, namely TA_Local_Opt_Col, to solve the problem using dynamic programming approach. Finally, a cool virtual submesh is formed by combining the local optimal cool virtual columns found in these areas. From this cool virtual submesh, a suitable-sized submesh with the lowest average temperature is allocated for J at one of its "corners".

References

[1]
I. Ababneh. An efficient free-list submesh allocation scheme for two-dimensional mesh-connected multicomputers. Journal of Systems and Software. (USA), 2006.
[2]
S. Bani-Mohammad, M. Ould-Khaoua and I. Ababneh An efficient non-contiguous processor allocation strategy for 2D mesh connected multicomputers. Information Sciences, Vol. 177, P. 2867--2883, 2007.
[3]
Intel Terascale Computing Research Program. http://www.intel.com/research/platform/terascale, 2008.
[4]
Zhijian Lu et al. Analysis of Temporal and Spatial Temperature Gradients for IC Reliability. TR CS-2004-08, University of Virginia, 2004.
[5]
S. Borkar Designing Reliable Systems from Unreliable Components: The Challenges of Transistor Variability and Degradation. IEEE Micro, vol. 25, n. 6, pp. 10--16 Nov.-Dec. 2005.
[6]
Zne-Jung Lee, Chou-Yuan Lee A hybrid search algorithm with heuristics for resource allocation problem. Journal of Information Sciences 173 (1-3) 155 -- 167, 2005.
[7]
G. De Micheli and L. Benini Networks on Chips: Technology and Tools. Morgan Kaufmann Publishers, 2006.
[8]
S.W. Chung and K. Skadron. Using On-Chip Event Counters for High-Resolution, Real-Time Temperature Measurement. IEEE/ASME ITHERM, 2006.
[9]
Krste Asanovic et al. The Landscape of Parallel Computing Research: A View from Berkeley. TR UCB/EECS-2006-183, UC Berkeley, 2006.
[10]
Chor Ping Low An Efficient Reconfiguration Algorithm for Degradable VLSI/WSI Arrays. IEEE Trans. on Computers, Vol.49, Iss.6, 2000.

Cited By

View all
  • (2021)Mapping techniques in multicore processors: current and future trendsThe Journal of Supercomputing10.1007/s11227-021-03650-6Online publication date: 5-Feb-2021
  • (2010)Hardware-enabled dynamic resource allocation for manycore systems using bidding-based system feedbackEURASIP Journal on Embedded Systems10.1155/2010/2614342010(1-21)Online publication date: 1-Jan-2010

Recommendations

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences
SPAA '08: Proceedings of the twentieth annual symposium on Parallelism in algorithms and architectures
June 2008
380 pages
ISBN:9781595939739
DOI:10.1145/1378533
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 14 June 2008

Permissions

Request permissions for this article.

Check for updates

Author Tags

  1. algorithm
  2. manycore chips
  3. noc
  4. submesh allocation
  5. temperature

Qualifiers

  • Research-article

Conference

SPAA08

Acceptance Rates

Overall Acceptance Rate 447 of 1,461 submissions, 31%

Upcoming Conference

SPAA '25
37th ACM Symposium on Parallelism in Algorithms and Architectures
July 28 - August 1, 2025
Portland , OR , USA

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)1
  • Downloads (Last 6 weeks)0
Reflects downloads up to 19 Feb 2025

Other Metrics

Citations

Cited By

View all
  • (2021)Mapping techniques in multicore processors: current and future trendsThe Journal of Supercomputing10.1007/s11227-021-03650-6Online publication date: 5-Feb-2021
  • (2010)Hardware-enabled dynamic resource allocation for manycore systems using bidding-based system feedbackEURASIP Journal on Embedded Systems10.1155/2010/2614342010(1-21)Online publication date: 1-Jan-2010

View Options

Login options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Figures

Tables

Media

Share

Share

Share this Publication link

Share on social media