Hardware and software support for NUMA computing on configurable emulated shared memory architectures

Martti Forsell, Erik Hansson, Christoph Kessler, Jari-Matti Mäkelä, Ville Leppänen

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

1 Citation (Scopus)

Abstract

The emulated shared memory (ESM) architectures are good candidates for future general purpose parallel computers due to their ability to provide easy-to-use explicitly parallel synchronous model of computation to programmers as well as avoid most performance bottlenecks present in current multicore architectures. In order to achieve full performance the applications must, however, have enough thread-level parallelism (TLP). To solve this problem, in our earlier work we have introduced a class of configurable emulated shared memory (CESM) machines that provides a special non-uniform memory access (NUMA) mode for situations where TLP is limited or for direct compatibility for legacy code sequential computing or NUMA mechanism. Unfortunately the earlier proposed CESM architecture does not integrate the different modes of the architecture well together e.g. by leaving the memories for different modes isolated and therefore the programming interface is non-integrated. In this paper we propose a number of hardware and software techniques to support NUMA computing in CESM architectures in a seamless way. The hardware techniques include three different NUMA-shared memory access mechanisms and the software ones provide a mechanism to integrate NUMA computation into the standard parallel random access machine (PRAM) operation of the CESM. The hardware techniques are evaluated on our REPLICA CESM architecture and compared to an ideal CESM machine making use of the proposed software techniques
Original languageEnglish
Title of host publication2013 IEEE International Symposium on Parallell & Distributed Processing, Workshops and Phd Forum
PublisherInstitute of Electrical and Electronic Engineers IEEE
Pages640-648
ISBN (Print)978-0-7695-4979-8
DOIs
Publication statusPublished - 2013
MoE publication typeA4 Article in a conference publication
EventIEEE 27th International Parallel and Distributed Processing Symposium Workshops and PhD Forum, IPDPSW 2013 - Boston, United States
Duration: 20 May 201324 May 2013
Conference number: 27

Conference

ConferenceIEEE 27th International Parallel and Distributed Processing Symposium Workshops and PhD Forum, IPDPSW 2013
Abbreviated titleIPDPSW 2013
CountryUnited States
CityBoston
Period20/05/1324/05/13

Fingerprint

Memory architecture
Computer hardware
Data storage equipment
Computer programming
Interfaces (computer)

Keywords

  • models of computation
  • NUMA
  • parallel computing
  • PRAM
  • programming model
  • Shared memory emulation

Cite this

Forsell, M., Hansson, E., Kessler, C., Mäkelä, J-M., & Leppänen, V. (2013). Hardware and software support for NUMA computing on configurable emulated shared memory architectures. In 2013 IEEE International Symposium on Parallell & Distributed Processing, Workshops and Phd Forum (pp. 640-648). Institute of Electrical and Electronic Engineers IEEE. https://doi.org/10.1109/IPDPSW.2013.146
Forsell, Martti ; Hansson, Erik ; Kessler, Christoph ; Mäkelä, Jari-Matti ; Leppänen, Ville. / Hardware and software support for NUMA computing on configurable emulated shared memory architectures. 2013 IEEE International Symposium on Parallell & Distributed Processing, Workshops and Phd Forum. Institute of Electrical and Electronic Engineers IEEE, 2013. pp. 640-648
@inproceedings{294486782bfd4eca8d3ef20259130a2e,
title = "Hardware and software support for NUMA computing on configurable emulated shared memory architectures",
abstract = "The emulated shared memory (ESM) architectures are good candidates for future general purpose parallel computers due to their ability to provide easy-to-use explicitly parallel synchronous model of computation to programmers as well as avoid most performance bottlenecks present in current multicore architectures. In order to achieve full performance the applications must, however, have enough thread-level parallelism (TLP). To solve this problem, in our earlier work we have introduced a class of configurable emulated shared memory (CESM) machines that provides a special non-uniform memory access (NUMA) mode for situations where TLP is limited or for direct compatibility for legacy code sequential computing or NUMA mechanism. Unfortunately the earlier proposed CESM architecture does not integrate the different modes of the architecture well together e.g. by leaving the memories for different modes isolated and therefore the programming interface is non-integrated. In this paper we propose a number of hardware and software techniques to support NUMA computing in CESM architectures in a seamless way. The hardware techniques include three different NUMA-shared memory access mechanisms and the software ones provide a mechanism to integrate NUMA computation into the standard parallel random access machine (PRAM) operation of the CESM. The hardware techniques are evaluated on our REPLICA CESM architecture and compared to an ideal CESM machine making use of the proposed software techniques",
keywords = "models of computation, NUMA, parallel computing, PRAM, programming model, Shared memory emulation",
author = "Martti Forsell and Erik Hansson and Christoph Kessler and Jari-Matti M{\"a}kel{\"a} and Ville Lepp{\"a}nen",
note = "CA2: TK810 SDA: ICT",
year = "2013",
doi = "10.1109/IPDPSW.2013.146",
language = "English",
isbn = "978-0-7695-4979-8",
pages = "640--648",
booktitle = "2013 IEEE International Symposium on Parallell & Distributed Processing, Workshops and Phd Forum",
publisher = "Institute of Electrical and Electronic Engineers IEEE",
address = "United States",

}

Forsell, M, Hansson, E, Kessler, C, Mäkelä, J-M & Leppänen, V 2013, Hardware and software support for NUMA computing on configurable emulated shared memory architectures. in 2013 IEEE International Symposium on Parallell & Distributed Processing, Workshops and Phd Forum. Institute of Electrical and Electronic Engineers IEEE, pp. 640-648, IEEE 27th International Parallel and Distributed Processing Symposium Workshops and PhD Forum, IPDPSW 2013, Boston, United States, 20/05/13. https://doi.org/10.1109/IPDPSW.2013.146

Hardware and software support for NUMA computing on configurable emulated shared memory architectures. / Forsell, Martti; Hansson, Erik; Kessler, Christoph; Mäkelä, Jari-Matti; Leppänen, Ville.

2013 IEEE International Symposium on Parallell & Distributed Processing, Workshops and Phd Forum. Institute of Electrical and Electronic Engineers IEEE, 2013. p. 640-648.

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - Hardware and software support for NUMA computing on configurable emulated shared memory architectures

AU - Forsell, Martti

AU - Hansson, Erik

AU - Kessler, Christoph

AU - Mäkelä, Jari-Matti

AU - Leppänen, Ville

N1 - CA2: TK810 SDA: ICT

PY - 2013

Y1 - 2013

N2 - The emulated shared memory (ESM) architectures are good candidates for future general purpose parallel computers due to their ability to provide easy-to-use explicitly parallel synchronous model of computation to programmers as well as avoid most performance bottlenecks present in current multicore architectures. In order to achieve full performance the applications must, however, have enough thread-level parallelism (TLP). To solve this problem, in our earlier work we have introduced a class of configurable emulated shared memory (CESM) machines that provides a special non-uniform memory access (NUMA) mode for situations where TLP is limited or for direct compatibility for legacy code sequential computing or NUMA mechanism. Unfortunately the earlier proposed CESM architecture does not integrate the different modes of the architecture well together e.g. by leaving the memories for different modes isolated and therefore the programming interface is non-integrated. In this paper we propose a number of hardware and software techniques to support NUMA computing in CESM architectures in a seamless way. The hardware techniques include three different NUMA-shared memory access mechanisms and the software ones provide a mechanism to integrate NUMA computation into the standard parallel random access machine (PRAM) operation of the CESM. The hardware techniques are evaluated on our REPLICA CESM architecture and compared to an ideal CESM machine making use of the proposed software techniques

AB - The emulated shared memory (ESM) architectures are good candidates for future general purpose parallel computers due to their ability to provide easy-to-use explicitly parallel synchronous model of computation to programmers as well as avoid most performance bottlenecks present in current multicore architectures. In order to achieve full performance the applications must, however, have enough thread-level parallelism (TLP). To solve this problem, in our earlier work we have introduced a class of configurable emulated shared memory (CESM) machines that provides a special non-uniform memory access (NUMA) mode for situations where TLP is limited or for direct compatibility for legacy code sequential computing or NUMA mechanism. Unfortunately the earlier proposed CESM architecture does not integrate the different modes of the architecture well together e.g. by leaving the memories for different modes isolated and therefore the programming interface is non-integrated. In this paper we propose a number of hardware and software techniques to support NUMA computing in CESM architectures in a seamless way. The hardware techniques include three different NUMA-shared memory access mechanisms and the software ones provide a mechanism to integrate NUMA computation into the standard parallel random access machine (PRAM) operation of the CESM. The hardware techniques are evaluated on our REPLICA CESM architecture and compared to an ideal CESM machine making use of the proposed software techniques

KW - models of computation

KW - NUMA

KW - parallel computing

KW - PRAM

KW - programming model

KW - Shared memory emulation

U2 - 10.1109/IPDPSW.2013.146

DO - 10.1109/IPDPSW.2013.146

M3 - Conference article in proceedings

SN - 978-0-7695-4979-8

SP - 640

EP - 648

BT - 2013 IEEE International Symposium on Parallell & Distributed Processing, Workshops and Phd Forum

PB - Institute of Electrical and Electronic Engineers IEEE

ER -

Forsell M, Hansson E, Kessler C, Mäkelä J-M, Leppänen V. Hardware and software support for NUMA computing on configurable emulated shared memory architectures. In 2013 IEEE International Symposium on Parallell & Distributed Processing, Workshops and Phd Forum. Institute of Electrical and Electronic Engineers IEEE. 2013. p. 640-648 https://doi.org/10.1109/IPDPSW.2013.146