Enhancing the Usability and Utilization of Accelerated Architectures via Docker

Nicholas Haydel, Sandra Gesing, Ian Taylor, Gregory Madey, Abdul Dakkak, Simon Garcia De Gonzalo, Wen-Mei W Hwu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Accelerated architectures such as GPUs (Graphics Processing Units) and MICs (Many Integrated Cores) have been proven to increase the performance of many algorithms compared to their CPU counterparts and are widely available in local, campus-wide and national infrastructures, however, their utilization is not following the same pace as their deployment. Reasons for the underutilization lay partly on the software side with proprietary and complex interfaces for development and usage. A common API providing an extra layer to abstract the differences and specific characteristics of those architectures would deliver a far more portable interface for application developers. This cloud challenge proposal presents such an API that addresses these issues using a container-based approach. The resulting environment provides Docker-based containers for deploying accelerator libraries, such as CUDA Toolkit, OpenCL and OpenACC, onto a wide variety of different platforms and operating systems. By leveraging the container approach, we can overlay accelerator libraries onto the host without needing to be concerned about the intricacies of underlying operating system of the host. Docker therefore provides the advantage of being easily applicable on diverse architectures, virtualizing the necessary environment and including libraries as well as applications in a standardized way. The novelty of our approach is the extra layer for utilization and device discovery in this layer improving the usability and uniform development of accelerated methods with direct access to resources.

Original languageEnglish (US)
Title of host publicationProceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015
EditorsOmer Rana, Rajkumar Buyya, Ioan Raicu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages361-367
Number of pages7
ISBN (Electronic)9780769556970
DOIs
StatePublished - Jan 1 2015
Event8th IEEE/ACM International Conference on Utility and Cloud Computing, UCC 2015 - Limassol, Cyprus
Duration: Dec 7 2015Dec 10 2015

Publication series

NameProceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015

Other

Other8th IEEE/ACM International Conference on Utility and Cloud Computing, UCC 2015
CountryCyprus
CityLimassol
Period12/7/1512/10/15

Fingerprint

Containers
Application programming interfaces (API)
Particle accelerators
Interfaces (computer)
Program processors
Graphics processing unit

Keywords

  • Docker
  • Virtualization
  • accelerated architectures
  • cloud computing
  • virtual machine

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

Haydel, N., Gesing, S., Taylor, I., Madey, G., Dakkak, A., De Gonzalo, S. G., & Hwu, W-M. W. (2015). Enhancing the Usability and Utilization of Accelerated Architectures via Docker. In O. Rana, R. Buyya, & I. Raicu (Eds.), Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015 (pp. 361-367). [7431432] (Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/UCC.2015.57

Enhancing the Usability and Utilization of Accelerated Architectures via Docker. / Haydel, Nicholas; Gesing, Sandra; Taylor, Ian; Madey, Gregory; Dakkak, Abdul; De Gonzalo, Simon Garcia; Hwu, Wen-Mei W.

Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015. ed. / Omer Rana; Rajkumar Buyya; Ioan Raicu. Institute of Electrical and Electronics Engineers Inc., 2015. p. 361-367 7431432 (Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Haydel, N, Gesing, S, Taylor, I, Madey, G, Dakkak, A, De Gonzalo, SG & Hwu, W-MW 2015, Enhancing the Usability and Utilization of Accelerated Architectures via Docker. in O Rana, R Buyya & I Raicu (eds), Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015., 7431432, Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015, Institute of Electrical and Electronics Engineers Inc., pp. 361-367, 8th IEEE/ACM International Conference on Utility and Cloud Computing, UCC 2015, Limassol, Cyprus, 12/7/15. https://doi.org/10.1109/UCC.2015.57
Haydel N, Gesing S, Taylor I, Madey G, Dakkak A, De Gonzalo SG et al. Enhancing the Usability and Utilization of Accelerated Architectures via Docker. In Rana O, Buyya R, Raicu I, editors, Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 361-367. 7431432. (Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015). https://doi.org/10.1109/UCC.2015.57
Haydel, Nicholas ; Gesing, Sandra ; Taylor, Ian ; Madey, Gregory ; Dakkak, Abdul ; De Gonzalo, Simon Garcia ; Hwu, Wen-Mei W. / Enhancing the Usability and Utilization of Accelerated Architectures via Docker. Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015. editor / Omer Rana ; Rajkumar Buyya ; Ioan Raicu. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 361-367 (Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015).
@inproceedings{0e09d7a733b3421b8e6b9d67b4e37b1d,
title = "Enhancing the Usability and Utilization of Accelerated Architectures via Docker",
abstract = "Accelerated architectures such as GPUs (Graphics Processing Units) and MICs (Many Integrated Cores) have been proven to increase the performance of many algorithms compared to their CPU counterparts and are widely available in local, campus-wide and national infrastructures, however, their utilization is not following the same pace as their deployment. Reasons for the underutilization lay partly on the software side with proprietary and complex interfaces for development and usage. A common API providing an extra layer to abstract the differences and specific characteristics of those architectures would deliver a far more portable interface for application developers. This cloud challenge proposal presents such an API that addresses these issues using a container-based approach. The resulting environment provides Docker-based containers for deploying accelerator libraries, such as CUDA Toolkit, OpenCL and OpenACC, onto a wide variety of different platforms and operating systems. By leveraging the container approach, we can overlay accelerator libraries onto the host without needing to be concerned about the intricacies of underlying operating system of the host. Docker therefore provides the advantage of being easily applicable on diverse architectures, virtualizing the necessary environment and including libraries as well as applications in a standardized way. The novelty of our approach is the extra layer for utilization and device discovery in this layer improving the usability and uniform development of accelerated methods with direct access to resources.",
keywords = "Docker, Virtualization, accelerated architectures, cloud computing, virtual machine",
author = "Nicholas Haydel and Sandra Gesing and Ian Taylor and Gregory Madey and Abdul Dakkak and {De Gonzalo}, {Simon Garcia} and Hwu, {Wen-Mei W}",
year = "2015",
month = "1",
day = "1",
doi = "10.1109/UCC.2015.57",
language = "English (US)",
series = "Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "361--367",
editor = "Omer Rana and Rajkumar Buyya and Ioan Raicu",
booktitle = "Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015",
address = "United States",

}

TY - GEN

T1 - Enhancing the Usability and Utilization of Accelerated Architectures via Docker

AU - Haydel, Nicholas

AU - Gesing, Sandra

AU - Taylor, Ian

AU - Madey, Gregory

AU - Dakkak, Abdul

AU - De Gonzalo, Simon Garcia

AU - Hwu, Wen-Mei W

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Accelerated architectures such as GPUs (Graphics Processing Units) and MICs (Many Integrated Cores) have been proven to increase the performance of many algorithms compared to their CPU counterparts and are widely available in local, campus-wide and national infrastructures, however, their utilization is not following the same pace as their deployment. Reasons for the underutilization lay partly on the software side with proprietary and complex interfaces for development and usage. A common API providing an extra layer to abstract the differences and specific characteristics of those architectures would deliver a far more portable interface for application developers. This cloud challenge proposal presents such an API that addresses these issues using a container-based approach. The resulting environment provides Docker-based containers for deploying accelerator libraries, such as CUDA Toolkit, OpenCL and OpenACC, onto a wide variety of different platforms and operating systems. By leveraging the container approach, we can overlay accelerator libraries onto the host without needing to be concerned about the intricacies of underlying operating system of the host. Docker therefore provides the advantage of being easily applicable on diverse architectures, virtualizing the necessary environment and including libraries as well as applications in a standardized way. The novelty of our approach is the extra layer for utilization and device discovery in this layer improving the usability and uniform development of accelerated methods with direct access to resources.

AB - Accelerated architectures such as GPUs (Graphics Processing Units) and MICs (Many Integrated Cores) have been proven to increase the performance of many algorithms compared to their CPU counterparts and are widely available in local, campus-wide and national infrastructures, however, their utilization is not following the same pace as their deployment. Reasons for the underutilization lay partly on the software side with proprietary and complex interfaces for development and usage. A common API providing an extra layer to abstract the differences and specific characteristics of those architectures would deliver a far more portable interface for application developers. This cloud challenge proposal presents such an API that addresses these issues using a container-based approach. The resulting environment provides Docker-based containers for deploying accelerator libraries, such as CUDA Toolkit, OpenCL and OpenACC, onto a wide variety of different platforms and operating systems. By leveraging the container approach, we can overlay accelerator libraries onto the host without needing to be concerned about the intricacies of underlying operating system of the host. Docker therefore provides the advantage of being easily applicable on diverse architectures, virtualizing the necessary environment and including libraries as well as applications in a standardized way. The novelty of our approach is the extra layer for utilization and device discovery in this layer improving the usability and uniform development of accelerated methods with direct access to resources.

KW - Docker

KW - Virtualization

KW - accelerated architectures

KW - cloud computing

KW - virtual machine

UR - http://www.scopus.com/inward/record.url?scp=84965031169&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84965031169&partnerID=8YFLogxK

U2 - 10.1109/UCC.2015.57

DO - 10.1109/UCC.2015.57

M3 - Conference contribution

AN - SCOPUS:84965031169

T3 - Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015

SP - 361

EP - 367

BT - Proceedings - 2015 IEEE/ACM 8th International Conference on Utility and Cloud Computing, UCC 2015

A2 - Rana, Omer

A2 - Buyya, Rajkumar

A2 - Raicu, Ioan

PB - Institute of Electrical and Electronics Engineers Inc.

ER -