Dynamic Application Reconfiguration on Heterogeneous HardwareCitation formats

Standard

Dynamic Application Reconfiguration on Heterogeneous Hardware. / Fumero, Juan; Papadimitriou, Michail; Zakkak, Foivos; Xekalaki, Maria; Clarkson, James; Kotselidis, Christos.

5th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE’19) . 2019. p. 165.

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

Harvard

Fumero, J, Papadimitriou, M, Zakkak, F, Xekalaki, M, Clarkson, J & Kotselidis, C 2019, Dynamic Application Reconfiguration on Heterogeneous Hardware. in 5th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE’19) . pp. 165. https://doi.org/10.1145/3313808.3313819

APA

Fumero, J., Papadimitriou, M., Zakkak, F., Xekalaki, M., Clarkson, J., & Kotselidis, C. (Accepted/In press). Dynamic Application Reconfiguration on Heterogeneous Hardware. In 5th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE’19) (pp. 165) https://doi.org/10.1145/3313808.3313819

Vancouver

Fumero J, Papadimitriou M, Zakkak F, Xekalaki M, Clarkson J, Kotselidis C. Dynamic Application Reconfiguration on Heterogeneous Hardware. In 5th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE’19) . 2019. p. 165 https://doi.org/10.1145/3313808.3313819

Author

Fumero, Juan ; Papadimitriou, Michail ; Zakkak, Foivos ; Xekalaki, Maria ; Clarkson, James ; Kotselidis, Christos. / Dynamic Application Reconfiguration on Heterogeneous Hardware. 5th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE’19) . 2019. pp. 165

Bibtex

@inproceedings{4296f1fe361d4781ad6791ec52290fc2,
title = "Dynamic Application Reconfiguration on Heterogeneous Hardware",
abstract = "By utilizing diverse heterogeneous hardware resources, developers can significantly improve the performance of their applications. Currently, in order to determine which parts of an application suit a particular type of hardware accelerator better, an offline analysis that uses a priori knowledge of the target hardware configuration is necessary. To make matters worse, the above process has to be repeated every time the application or the hardware configuration changes.This paper introduces TornadoVM, a virtual machine capable of reconfiguring applications, at run-time, for hardware acceleration based on the currently available hardware re- sources. Through TornadoVM, we introduce a new level of compilation in which applications can benefit from heterogeneous hardware. We showcase the capabilities of TornadoVM by executing a complex computer vision application and six benchmarks on a heterogeneous system that includes a CPU, an FPGA, and a GPU. Our evaluation shows that by using dynamic reconfiguration, we achieve an average of 7.7× speedup over the statically-configured accelerated code.",
author = "Juan Fumero and Michail Papadimitriou and Foivos Zakkak and Maria Xekalaki and James Clarkson and Christos Kotselidis",
year = "2019",
month = "4",
day = "14",
doi = "10.1145/3313808.3313819",
language = "English",
pages = "165",
booktitle = "5th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE’19)",

}

RIS

TY - GEN

T1 - Dynamic Application Reconfiguration on Heterogeneous Hardware

AU - Fumero, Juan

AU - Papadimitriou, Michail

AU - Zakkak, Foivos

AU - Xekalaki, Maria

AU - Clarkson, James

AU - Kotselidis, Christos

PY - 2019/4/14

Y1 - 2019/4/14

N2 - By utilizing diverse heterogeneous hardware resources, developers can significantly improve the performance of their applications. Currently, in order to determine which parts of an application suit a particular type of hardware accelerator better, an offline analysis that uses a priori knowledge of the target hardware configuration is necessary. To make matters worse, the above process has to be repeated every time the application or the hardware configuration changes.This paper introduces TornadoVM, a virtual machine capable of reconfiguring applications, at run-time, for hardware acceleration based on the currently available hardware re- sources. Through TornadoVM, we introduce a new level of compilation in which applications can benefit from heterogeneous hardware. We showcase the capabilities of TornadoVM by executing a complex computer vision application and six benchmarks on a heterogeneous system that includes a CPU, an FPGA, and a GPU. Our evaluation shows that by using dynamic reconfiguration, we achieve an average of 7.7× speedup over the statically-configured accelerated code.

AB - By utilizing diverse heterogeneous hardware resources, developers can significantly improve the performance of their applications. Currently, in order to determine which parts of an application suit a particular type of hardware accelerator better, an offline analysis that uses a priori knowledge of the target hardware configuration is necessary. To make matters worse, the above process has to be repeated every time the application or the hardware configuration changes.This paper introduces TornadoVM, a virtual machine capable of reconfiguring applications, at run-time, for hardware acceleration based on the currently available hardware re- sources. Through TornadoVM, we introduce a new level of compilation in which applications can benefit from heterogeneous hardware. We showcase the capabilities of TornadoVM by executing a complex computer vision application and six benchmarks on a heterogeneous system that includes a CPU, an FPGA, and a GPU. Our evaluation shows that by using dynamic reconfiguration, we achieve an average of 7.7× speedup over the statically-configured accelerated code.

U2 - 10.1145/3313808.3313819

DO - 10.1145/3313808.3313819

M3 - Conference contribution

SP - 165

BT - 5th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE’19)

ER -