Distributed event-based computing
Distributed event-based computing
As computing systems get larger in capability-a good thing-they also get larger in ways less desirable: Cost, volume, power requirements and so on. Further, as the datastructures necessary to support large computations grow physically, the proportion of wallclock time spent communicating increases dramatically at the expense of the time spent calculating. This state of affairs is currently unacceptable and will only get worse as exa-scale machines move from the esoteric to the commonplace. As the unit cost of non-trivial cores continues to fall, one powerful approach is to build systems that have immense numbers of relatively small cores embedded (both geometrically and topologically) in a vast distributed network of stored state data: Take the compute to the data, rather than the other way round. In this paper, we describe POETS-Partially Ordered Event Triggered Systems. This is a novel kind of computing architecture, built upon the neuromorphic concept that has inspired such machines as SpiNNaker[1,2] and BrainScaleS[3]. The central idea is that a problem is broken down into a large set of interacting devices, which communicate asynchronously via small, hardware brokered packets (the arrival of which is an event). The set of devices is the task graph. You cannot take a conventional codebase and port it to a POETS architecture; it is necessary to strip the application back to the underlying mathematics and reconstruct the algorithm in a manner sympathetic to the solution capabilities of the machine. However, for the class of problems for which this approach is suitable, POETS has already demonstrated solution speedups of a factor of 200 over conventional techniques.
accelerators, heterogeneous systems, Multicore/manycore systems
583-592
Brown, Andrew
5c19e523-65ec-499b-9e7c-91522017d7e0
Thomas, David
45ca4f53-8eae-411f-9824-af94b0d291cb
Reeve, Jeff
dd909010-7d44-44ea-83fe-a09e4d492618
Tarawneh, Ghaith
1b90fbe9-1337-4216-83ba-b01115bf4000
De Gennaro, Alessandro
8c78f093-4a6b-4e21-90cc-fb90e4a1c8b3
Mokhov, Andrey
7ad0909b-34e8-4f32-908c-b6406b397776
Naylor, Matthew
7f0b28f7-b50f-40cc-b3ed-3821e5ef1230
Kazmierski, Tom
a97d7958-40c3-413f-924d-84545216092a
1 March 2018
Brown, Andrew
5c19e523-65ec-499b-9e7c-91522017d7e0
Thomas, David
45ca4f53-8eae-411f-9824-af94b0d291cb
Reeve, Jeff
dd909010-7d44-44ea-83fe-a09e4d492618
Tarawneh, Ghaith
1b90fbe9-1337-4216-83ba-b01115bf4000
De Gennaro, Alessandro
8c78f093-4a6b-4e21-90cc-fb90e4a1c8b3
Mokhov, Andrey
7ad0909b-34e8-4f32-908c-b6406b397776
Naylor, Matthew
7f0b28f7-b50f-40cc-b3ed-3821e5ef1230
Kazmierski, Tom
a97d7958-40c3-413f-924d-84545216092a
Brown, Andrew, Thomas, David, Reeve, Jeff, Tarawneh, Ghaith, De Gennaro, Alessandro, Mokhov, Andrey, Naylor, Matthew and Kazmierski, Tom
(2018)
Distributed event-based computing.
In Parallel Computing is Everywhere: Parallel Computing is Everywhere.
vol. 32,
IOS Press.
.
(doi:10.3233/978-1-61499-843-3-583).
Record type:
Conference or Workshop Item
(Paper)
Abstract
As computing systems get larger in capability-a good thing-they also get larger in ways less desirable: Cost, volume, power requirements and so on. Further, as the datastructures necessary to support large computations grow physically, the proportion of wallclock time spent communicating increases dramatically at the expense of the time spent calculating. This state of affairs is currently unacceptable and will only get worse as exa-scale machines move from the esoteric to the commonplace. As the unit cost of non-trivial cores continues to fall, one powerful approach is to build systems that have immense numbers of relatively small cores embedded (both geometrically and topologically) in a vast distributed network of stored state data: Take the compute to the data, rather than the other way round. In this paper, we describe POETS-Partially Ordered Event Triggered Systems. This is a novel kind of computing architecture, built upon the neuromorphic concept that has inspired such machines as SpiNNaker[1,2] and BrainScaleS[3]. The central idea is that a problem is broken down into a large set of interacting devices, which communicate asynchronously via small, hardware brokered packets (the arrival of which is an event). The set of devices is the task graph. You cannot take a conventional codebase and port it to a POETS architecture; it is necessary to strip the application back to the underlying mathematics and reconstruct the algorithm in a manner sympathetic to the solution capabilities of the machine. However, for the class of problems for which this approach is suitable, POETS has already demonstrated solution speedups of a factor of 200 over conventional techniques.
Text
Distributed event-based computing
- Accepted Manuscript
More information
Accepted/In Press date: 14 August 2017
e-pub ahead of print date: 1 March 2018
Published date: 1 March 2018
Additional Information:
This paper will be published in the conf proceedings, they have not been published yet.
Venue - Dates:
ParCo, , Bolonga, Italy, 2017-09-12 - 2017-09-15
Keywords:
accelerators, heterogeneous systems, Multicore/manycore systems
Identifiers
Local EPrints ID: 418177
URI: http://eprints.soton.ac.uk/id/eprint/418177
ISSN: 0927-5452
PURE UUID: 9fd04853-753c-4ede-8e9c-9091744affe5
Catalogue record
Date deposited: 23 Feb 2018 17:30
Last modified: 15 Mar 2024 18:34
Export record
Altmetrics
Contributors
Author:
Andrew Brown
Author:
David Thomas
Author:
Jeff Reeve
Author:
Ghaith Tarawneh
Author:
Alessandro De Gennaro
Author:
Andrey Mokhov
Author:
Matthew Naylor
Author:
Tom Kazmierski
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