Advanced

A comparative study of scheduling techniques for multimedia applications on SIMD pipelines

Arslan, Mehmet Ali LU ; Gruian, Flavius LU and Kuchcinski, Krzysztof LU (2015) DATE Friday Workshop on Heterogeneous Architectures and Design Methods for Embedded Image Systems (HIS 2015) In DATE Friday Workshop on Heterogeneous Architectures and Design Methods for Embedded Image Systems (HIS 2015) p.3-9
Abstract
Parallel architectures are essential in order to take advantage of the parallelism inherent in streaming applications. One particular branch of these employ hardware SIMD pipelines. In this paper, we analyse several scheduling techniques, namely ad hoc overlapped execution, modulo scheduling and modulo scheduling with unrolling, all of which aim to efficiently utilize the special architecture design. Our investigation focuses on improving throughput while analysing other metrics that are important for streaming applications, such as register pressure, buffer sizes and code size. Through experiments conducted on several media benchmarks, we present and discuss trade-offs involved when selecting any one of these scheduling techniques.
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
in
DATE Friday Workshop on Heterogeneous Architectures and Design Methods for Embedded Image Systems (HIS 2015)
editor
Hannig, Frank; Fey, Dietmar; Lokhmotov, Anton; ; and
pages
7 pages
conference name
DATE Friday Workshop on Heterogeneous Architectures and Design Methods for Embedded Image Systems (HIS 2015)
ISBN
978-3-9815370-4-8
project
HiPEC
language
English
LU publication?
yes
id
277b29c0-5535-4374-abe9-59e3a2d52f2b (old id 5153964)
alternative location
http://arxiv.org/pdf/1502.07447.pdf
date added to LUP
2015-03-13 10:36:42
date last changed
2016-08-24 11:31:39
@inproceedings{277b29c0-5535-4374-abe9-59e3a2d52f2b,
  abstract     = {Parallel architectures are essential in order to take advantage of the parallelism inherent in streaming applications. One particular branch of these employ hardware SIMD pipelines. In this paper, we analyse several scheduling techniques, namely ad hoc overlapped execution, modulo scheduling and modulo scheduling with unrolling, all of which aim to efficiently utilize the special architecture design. Our investigation focuses on improving throughput while analysing other metrics that are important for streaming applications, such as register pressure, buffer sizes and code size. Through experiments conducted on several media benchmarks, we present and discuss trade-offs involved when selecting any one of these scheduling techniques.},
  author       = {Arslan, Mehmet Ali and Gruian, Flavius and Kuchcinski, Krzysztof},
  booktitle    = {DATE Friday Workshop on Heterogeneous Architectures and Design Methods for Embedded Image Systems (HIS 2015)},
  editor       = {Hannig, Frank and Fey, Dietmar and Lokhmotov, Anton},
  isbn         = {978-3-9815370-4-8},
  language     = {eng},
  pages        = {3--9},
  title        = {A comparative study of scheduling techniques for multimedia applications on SIMD pipelines},
  year         = {2015},
}