Adaptive resource allocation for streaming applications

T.D. ter Braak, Hermen Toersche, Andre B.J. Kokkeler, Gerardus Johannes Maria Smit

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

  • 8 Citations

Abstract

Streaming applications often have latency and throughput requirements due to timing critical signal processing, or the time critical interaction with their environment. Mapping such applications to a multi-core architecture is commonly done at design-time to be able to analyze the complex design-space. However, such design-flows cannot deal with a dynamic platform or a dynamic set of applications. Hardware faults and resources claimed by other applications may render the assumed available resources inaccessible. To avoid the assumptions posed on the state of the platform by a fixed resource allocation, applica- tions should be designed with location transparency in mind. Applications must be analyzed at design-time to determine the required resource budget, independent of which specific resources will be allocated. Sufficient performance can be guaranteed when such applications are mapped onto an architecture in which each resource is arbitrated using a budget scheduler. Within the Cutting edge Reconfigurable ICs for Stream Processing (CRISP) project, a many-core platform is developed that adheres to these requirements. Using the configuration features of the platform, the system is able to control at run-time what resources are being used by the applications. This paper shows that run-time resource allocation can effectively adapt to the available set of resources, providing partial distribution transparency to the user. As an example, a GNSS receiver is mapped to the platform containing faulty hardware components. A few resources remain critical, but in most cases the faulty components can be circumvented, such that adequate resources can be allocated to the application at run-time.
LanguageUndefined
Title of host publicationInternational Conference on Embedded Computer Systems: Architectures, Modeling and Simulation, IC-SAMOS 2011
EditorsL. Carro, A.D. Pimentel
Place of PublicationUSA
PublisherIEEE Circuits & Systems Society
Pages388-395
Number of pages8
ISBN (Print)978-1-4577-0800-8
DOIs
StatePublished - 18 Jul 2011
Event2011 International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation, IC-SAMOS XI - Institute of East Aegean, Samos, Greece
Duration: 18 Jul 201121 Jul 2011
Conference number: 11

Publication series

Name
PublisherIEEE Circuits & Systems Society

Conference

Conference2011 International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation, IC-SAMOS XI
Abbreviated titleIC-SAMOS
CountryGreece
CitySamos
Period18/07/1121/07/11

Keywords

  • METIS-278732
  • EWI-20323
  • EC Grant Agreement nr.: FP7/215881
  • IR-77923

Cite this

ter Braak, T. D., Toersche, H., Kokkeler, A. B. J., & Smit, G. J. M. (2011). Adaptive resource allocation for streaming applications. In L. Carro, & A. D. Pimentel (Eds.), International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation, IC-SAMOS 2011 (pp. 388-395). USA: IEEE Circuits & Systems Society. DOI: 10.1109/SAMOS.2011.6045489
ter Braak, T.D. ; Toersche, Hermen ; Kokkeler, Andre B.J. ; Smit, Gerardus Johannes Maria. / Adaptive resource allocation for streaming applications. International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation, IC-SAMOS 2011. editor / L. Carro ; A.D. Pimentel. USA : IEEE Circuits & Systems Society, 2011. pp. 388-395
@inproceedings{4ff7747800b9451aa1d313350daf4696,
title = "Adaptive resource allocation for streaming applications",
abstract = "Streaming applications often have latency and throughput requirements due to timing critical signal processing, or the time critical interaction with their environment. Mapping such applications to a multi-core architecture is commonly done at design-time to be able to analyze the complex design-space. However, such design-flows cannot deal with a dynamic platform or a dynamic set of applications. Hardware faults and resources claimed by other applications may render the assumed available resources inaccessible. To avoid the assumptions posed on the state of the platform by a fixed resource allocation, applica- tions should be designed with location transparency in mind. Applications must be analyzed at design-time to determine the required resource budget, independent of which specific resources will be allocated. Sufficient performance can be guaranteed when such applications are mapped onto an architecture in which each resource is arbitrated using a budget scheduler. Within the Cutting edge Reconfigurable ICs for Stream Processing (CRISP) project, a many-core platform is developed that adheres to these requirements. Using the configuration features of the platform, the system is able to control at run-time what resources are being used by the applications. This paper shows that run-time resource allocation can effectively adapt to the available set of resources, providing partial distribution transparency to the user. As an example, a GNSS receiver is mapped to the platform containing faulty hardware components. A few resources remain critical, but in most cases the faulty components can be circumvented, such that adequate resources can be allocated to the application at run-time.",
keywords = "METIS-278732, EWI-20323, EC Grant Agreement nr.: FP7/215881, IR-77923",
author = "{ter Braak}, T.D. and Hermen Toersche and Kokkeler, {Andre B.J.} and Smit, {Gerardus Johannes Maria}",
note = "eemcs-eprint-20323",
year = "2011",
month = "7",
day = "18",
doi = "10.1109/SAMOS.2011.6045489",
language = "Undefined",
isbn = "978-1-4577-0800-8",
publisher = "IEEE Circuits & Systems Society",
pages = "388--395",
editor = "L. Carro and A.D. Pimentel",
booktitle = "International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation, IC-SAMOS 2011",

}

ter Braak, TD, Toersche, H, Kokkeler, ABJ & Smit, GJM 2011, Adaptive resource allocation for streaming applications. in L Carro & AD Pimentel (eds), International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation, IC-SAMOS 2011. IEEE Circuits & Systems Society, USA, pp. 388-395, 2011 International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation, IC-SAMOS XI, Samos, Greece, 18/07/11. DOI: 10.1109/SAMOS.2011.6045489

Adaptive resource allocation for streaming applications. / ter Braak, T.D.; Toersche, Hermen; Kokkeler, Andre B.J.; Smit, Gerardus Johannes Maria.

International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation, IC-SAMOS 2011. ed. / L. Carro; A.D. Pimentel. USA : IEEE Circuits & Systems Society, 2011. p. 388-395.

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

TY - GEN

T1 - Adaptive resource allocation for streaming applications

AU - ter Braak,T.D.

AU - Toersche,Hermen

AU - Kokkeler,Andre B.J.

AU - Smit,Gerardus Johannes Maria

N1 - eemcs-eprint-20323

PY - 2011/7/18

Y1 - 2011/7/18

N2 - Streaming applications often have latency and throughput requirements due to timing critical signal processing, or the time critical interaction with their environment. Mapping such applications to a multi-core architecture is commonly done at design-time to be able to analyze the complex design-space. However, such design-flows cannot deal with a dynamic platform or a dynamic set of applications. Hardware faults and resources claimed by other applications may render the assumed available resources inaccessible. To avoid the assumptions posed on the state of the platform by a fixed resource allocation, applica- tions should be designed with location transparency in mind. Applications must be analyzed at design-time to determine the required resource budget, independent of which specific resources will be allocated. Sufficient performance can be guaranteed when such applications are mapped onto an architecture in which each resource is arbitrated using a budget scheduler. Within the Cutting edge Reconfigurable ICs for Stream Processing (CRISP) project, a many-core platform is developed that adheres to these requirements. Using the configuration features of the platform, the system is able to control at run-time what resources are being used by the applications. This paper shows that run-time resource allocation can effectively adapt to the available set of resources, providing partial distribution transparency to the user. As an example, a GNSS receiver is mapped to the platform containing faulty hardware components. A few resources remain critical, but in most cases the faulty components can be circumvented, such that adequate resources can be allocated to the application at run-time.

AB - Streaming applications often have latency and throughput requirements due to timing critical signal processing, or the time critical interaction with their environment. Mapping such applications to a multi-core architecture is commonly done at design-time to be able to analyze the complex design-space. However, such design-flows cannot deal with a dynamic platform or a dynamic set of applications. Hardware faults and resources claimed by other applications may render the assumed available resources inaccessible. To avoid the assumptions posed on the state of the platform by a fixed resource allocation, applica- tions should be designed with location transparency in mind. Applications must be analyzed at design-time to determine the required resource budget, independent of which specific resources will be allocated. Sufficient performance can be guaranteed when such applications are mapped onto an architecture in which each resource is arbitrated using a budget scheduler. Within the Cutting edge Reconfigurable ICs for Stream Processing (CRISP) project, a many-core platform is developed that adheres to these requirements. Using the configuration features of the platform, the system is able to control at run-time what resources are being used by the applications. This paper shows that run-time resource allocation can effectively adapt to the available set of resources, providing partial distribution transparency to the user. As an example, a GNSS receiver is mapped to the platform containing faulty hardware components. A few resources remain critical, but in most cases the faulty components can be circumvented, such that adequate resources can be allocated to the application at run-time.

KW - METIS-278732

KW - EWI-20323

KW - EC Grant Agreement nr.: FP7/215881

KW - IR-77923

U2 - 10.1109/SAMOS.2011.6045489

DO - 10.1109/SAMOS.2011.6045489

M3 - Conference contribution

SN - 978-1-4577-0800-8

SP - 388

EP - 395

BT - International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation, IC-SAMOS 2011

PB - IEEE Circuits & Systems Society

CY - USA

ER -

ter Braak TD, Toersche H, Kokkeler ABJ, Smit GJM. Adaptive resource allocation for streaming applications. In Carro L, Pimentel AD, editors, International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation, IC-SAMOS 2011. USA: IEEE Circuits & Systems Society. 2011. p. 388-395. Available from, DOI: 10.1109/SAMOS.2011.6045489