Just-in-time scheduling techniques for multicore signal processing systems

Julien Heulot; Maxime Pelcat; Jean Francois Nezan; Yaset Oliva; Slaheddine Aridhi; Shuvra S. Bhattacharyya

doi:10.1109/GlobalSIP.2014.7032071

Just-in-time scheduling techniques for multicore signal processing systems

Julien Heulot
, Maxime Pelcat
, Jean Francois Nezan
, Yaset Oliva
, Slaheddine Aridhi
, Shuvra S. Bhattacharyya

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

3 Citations (Scopus)

Abstract

This paper introduces a novel multicore scheduling method that leverages a parameterized dataflow Model of Computation (MoC). This method, which we have named Just-In-Time Multicore Scheduling (JIT-MS), aims to efficiently schedule Parameterized and Interfaced Synchronous DataFlow (PiSDF) graphs on multicore architectures. This method exploits features of PiSDF to And locally static regions that exhibit predictable communications. This paper uses a multicore signal processing benchmark to demonstrate that the JIT-MS scheduler can exploit more parallelism than a conventional multicore task scheduler based on task creation and dispatch. Experimental results of the JIT-MS on an 8-core Texas Instruments Keystone Digital Signal Processor (DSP) are compared with those obtained from the OpenMP implementation provided by Texas Instruments. Results shows latency improvements of up to 26% for multicore signal processing systems.

Original language	English
Title of host publication	2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014
Publisher	IEEE
Pages	25-29
Number of pages	5
ISBN (Electronic)	9781479970889
DOIs	https://doi.org/10.1109/GlobalSIP.2014.7032071
Publication status	Published - 5 Feb 2014
Publication type	A4 Article in conference proceedings
Event	2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014 - Atlanta, United States Duration: 3 Dec 2014 → 5 Dec 2014

Conference

Conference	2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014
Country/Territory	United States
City	Atlanta
Period	3/12/14 → 5/12/14

ASJC Scopus subject areas

Signal Processing
Information Systems

Access to Document

10.1109/GlobalSIP.2014.7032071

Cite this

@inproceedings{d5ba3eb2792646f5b34554f636e3ffb8,

title = "Just-in-time scheduling techniques for multicore signal processing systems",

abstract = "This paper introduces a novel multicore scheduling method that leverages a parameterized dataflow Model of Computation (MoC). This method, which we have named Just-In-Time Multicore Scheduling (JIT-MS), aims to efficiently schedule Parameterized and Interfaced Synchronous DataFlow (PiSDF) graphs on multicore architectures. This method exploits features of PiSDF to And locally static regions that exhibit predictable communications. This paper uses a multicore signal processing benchmark to demonstrate that the JIT-MS scheduler can exploit more parallelism than a conventional multicore task scheduler based on task creation and dispatch. Experimental results of the JIT-MS on an 8-core Texas Instruments Keystone Digital Signal Processor (DSP) are compared with those obtained from the OpenMP implementation provided by Texas Instruments. Results shows latency improvements of up to 26\% for multicore signal processing systems.",

author = "Julien Heulot and Maxime Pelcat and Nezan, \{Jean Francois\} and Yaset Oliva and Slaheddine Aridhi and Bhattacharyya, \{Shuvra S.\}",

year = "2014",

month = feb,

day = "5",

doi = "10.1109/GlobalSIP.2014.7032071",

language = "English",

pages = "25--29",

booktitle = "2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014",

publisher = "IEEE",

address = "United States",

note = "2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014 ; Conference date: 03-12-2014 Through 05-12-2014",

}

Heulot, J, Pelcat, M, Nezan, JF, Oliva, Y, Aridhi, S & Bhattacharyya, SS 2014, Just-in-time scheduling techniques for multicore signal processing systems. in 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014., 7032071, IEEE, pp. 25-29, 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014, Atlanta, United States, 3/12/14. https://doi.org/10.1109/GlobalSIP.2014.7032071

TY - GEN

T1 - Just-in-time scheduling techniques for multicore signal processing systems

AU - Heulot, Julien

AU - Pelcat, Maxime

AU - Nezan, Jean Francois

AU - Oliva, Yaset

AU - Aridhi, Slaheddine

AU - Bhattacharyya, Shuvra S.

PY - 2014/2/5

Y1 - 2014/2/5

N2 - This paper introduces a novel multicore scheduling method that leverages a parameterized dataflow Model of Computation (MoC). This method, which we have named Just-In-Time Multicore Scheduling (JIT-MS), aims to efficiently schedule Parameterized and Interfaced Synchronous DataFlow (PiSDF) graphs on multicore architectures. This method exploits features of PiSDF to And locally static regions that exhibit predictable communications. This paper uses a multicore signal processing benchmark to demonstrate that the JIT-MS scheduler can exploit more parallelism than a conventional multicore task scheduler based on task creation and dispatch. Experimental results of the JIT-MS on an 8-core Texas Instruments Keystone Digital Signal Processor (DSP) are compared with those obtained from the OpenMP implementation provided by Texas Instruments. Results shows latency improvements of up to 26% for multicore signal processing systems.

AB - This paper introduces a novel multicore scheduling method that leverages a parameterized dataflow Model of Computation (MoC). This method, which we have named Just-In-Time Multicore Scheduling (JIT-MS), aims to efficiently schedule Parameterized and Interfaced Synchronous DataFlow (PiSDF) graphs on multicore architectures. This method exploits features of PiSDF to And locally static regions that exhibit predictable communications. This paper uses a multicore signal processing benchmark to demonstrate that the JIT-MS scheduler can exploit more parallelism than a conventional multicore task scheduler based on task creation and dispatch. Experimental results of the JIT-MS on an 8-core Texas Instruments Keystone Digital Signal Processor (DSP) are compared with those obtained from the OpenMP implementation provided by Texas Instruments. Results shows latency improvements of up to 26% for multicore signal processing systems.

UR - https://www.scopus.com/pages/publications/84974558043

U2 - 10.1109/GlobalSIP.2014.7032071

DO - 10.1109/GlobalSIP.2014.7032071

M3 - Conference contribution

AN - SCOPUS:84974558043

SP - 25

EP - 29

BT - 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014

PB - IEEE

T2 - 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014

Y2 - 3 December 2014 through 5 December 2014

ER -

Just-in-time scheduling techniques for multicore signal processing systems

Abstract

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this