The University of Southampton
University of Southampton Institutional Repository

Synthesis of multi-FPGA systems with asynchronous communications

Synthesis of multi-FPGA systems with asynchronous communications
Synthesis of multi-FPGA systems with asynchronous communications

MOODS (Multiple Objective Optimisation in Data and control path Synthesis) is a high-level synthesis system which provides the ability to synthesise a system level behavioural description into a structural representation.  This thesis describes an enhancement to the original MOODS system that provides an automated mechanism to target a single behavioural input design onto heterogeneous re-configurable devices, forming a multi-FPGA system.  This thesis focuses on some of the problems associated with multi-FPGA synthesis, in particular the area utilisation of target devices and input/output (I/O) constraints in a multi-FGPA system.

The multi-FPGA partitioning mechanism has added a new optimisation objective into the MOODS synthesis system.  Not only does it provide an automated means of partitioning the design into separate blocks, the partitioning algorithm optimises the utilisation of device area and I/O taking into account the activity profile of the design, and allows performance and I/O utilisation trade-offs to be considered.  Asynchronous channel-based communication and pipelining techniques in multi-FPGA synthesis can produce a multi-FPGA system with performance close to a single-device implementation.

Results showed that the multi-FPGA synthesis enhancement integrated within the MOODS environment provided a rapid realisation of pipelined multi-FPGA systems with asynchronous communication channels at the expense of an acceptable increase in area overhead and design latency.

University of Southampton
Yee, Tack Boon
f9d970a9-5346-4fd8-b5bb-aa502aa14340
Yee, Tack Boon
f9d970a9-5346-4fd8-b5bb-aa502aa14340

Yee, Tack Boon (2007) Synthesis of multi-FPGA systems with asynchronous communications. University of Southampton, Doctoral Thesis.

Record type: Thesis (Doctoral)

Abstract

MOODS (Multiple Objective Optimisation in Data and control path Synthesis) is a high-level synthesis system which provides the ability to synthesise a system level behavioural description into a structural representation.  This thesis describes an enhancement to the original MOODS system that provides an automated mechanism to target a single behavioural input design onto heterogeneous re-configurable devices, forming a multi-FPGA system.  This thesis focuses on some of the problems associated with multi-FPGA synthesis, in particular the area utilisation of target devices and input/output (I/O) constraints in a multi-FGPA system.

The multi-FPGA partitioning mechanism has added a new optimisation objective into the MOODS synthesis system.  Not only does it provide an automated means of partitioning the design into separate blocks, the partitioning algorithm optimises the utilisation of device area and I/O taking into account the activity profile of the design, and allows performance and I/O utilisation trade-offs to be considered.  Asynchronous channel-based communication and pipelining techniques in multi-FPGA synthesis can produce a multi-FPGA system with performance close to a single-device implementation.

Results showed that the multi-FPGA synthesis enhancement integrated within the MOODS environment provided a rapid realisation of pipelined multi-FPGA systems with asynchronous communication channels at the expense of an acceptable increase in area overhead and design latency.

Text
1041449_v.2.pdf - Version of Record
Available under License University of Southampton Thesis Licence.
Download (17MB)
Text
1041449_v.1.pdf - Version of Record
Available under License University of Southampton Thesis Licence.
Download (18MB)
Archive
1041449_v.1_data.zip - Version of Record
Available under License University of Southampton Thesis Licence.
Download (65MB)

More information

Published date: 2007

Identifiers

Local EPrints ID: 466117
URI: http://eprints.soton.ac.uk/id/eprint/466117
PURE UUID: 27cc5d15-706e-461b-ad4b-aad985ef326c

Catalogue record

Date deposited: 05 Jul 2022 04:24
Last modified: 16 Mar 2024 20:31

Export record

Contributors

Author: Tack Boon Yee

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×