Scalability and robustness of a market-based network resource allocation system
Scalability and robustness of a market-based network resource allocation system
In this paper, we consider issues related to scalability and robustness in designing a market-based multi-agent system that allocates bandwidth in a communications network. Specifically, an empirical evaluation is carried out to assess the system performance under a variety of design configurations in order to provide an insight into network deployment issues. This extends our previous work in which we developed an application that makes use of market-based software agents that compete in decentralised marketplaces to buy and sell bandwidth resources. Our new results show that given a light to moderate network traffic load, partitioning the network into a few regions, each with a separate market server, gives a higher call success rate than by using a single market. Moreover, a trade-off in the number of regions was also noted between the average call success rate and the number of messages received per market server. Finally, given the possibility of market failures, we observe that the average call success rates increase with an increasing number of markets until a maximum is reached.
Agent, Market, Resource allocation, Communications network, Scalability, Robustness
2-913923-14-3
30-39
Haque, Nadim
e64efd61-9b59-412a-825e-cdede3a1e6c3
Jennings, N. R.
ab3d94cc-247c-4545-9d1e-65873d6cdb30
Moreau, Luc
033c63dd-3fe9-4040-849f-dfccbe0406f8
Grosu, Daniel
b195c243-4b3a-4166-8df9-afc9b70c1569
Shapiro, Jonathan
3f516d95-f000-4646-a3d9-09017ee20da1
2005
Haque, Nadim
e64efd61-9b59-412a-825e-cdede3a1e6c3
Jennings, N. R.
ab3d94cc-247c-4545-9d1e-65873d6cdb30
Moreau, Luc
033c63dd-3fe9-4040-849f-dfccbe0406f8
Grosu, Daniel
b195c243-4b3a-4166-8df9-afc9b70c1569
Shapiro, Jonathan
3f516d95-f000-4646-a3d9-09017ee20da1
Haque, Nadim, Jennings, N. R. and Moreau, Luc
(2005)
Scalability and robustness of a market-based network resource allocation system.
Grosu, Daniel and Shapiro, Jonathan
(eds.)
The First International Workshop on Incentive Based Computing (IBC'05), University of Technology of Compiegne, France.
.
Record type:
Conference or Workshop Item
(Paper)
Abstract
In this paper, we consider issues related to scalability and robustness in designing a market-based multi-agent system that allocates bandwidth in a communications network. Specifically, an empirical evaluation is carried out to assess the system performance under a variety of design configurations in order to provide an insight into network deployment issues. This extends our previous work in which we developed an application that makes use of market-based software agents that compete in decentralised marketplaces to buy and sell bandwidth resources. Our new results show that given a light to moderate network traffic load, partitioning the network into a few regions, each with a separate market server, gives a higher call success rate than by using a single market. Moreover, a trade-off in the number of regions was also noted between the average call success rate and the number of messages received per market server. Finally, given the possibility of market failures, we observe that the average call success rates increase with an increasing number of markets until a maximum is reached.
Text
netnomics06.pdf
- Accepted Manuscript
More information
Published date: 2005
Additional Information:
Event Dates: 19th September 2005
Venue - Dates:
The First International Workshop on Incentive Based Computing (IBC'05), University of Technology of Compiegne, France, 2005-09-19
Keywords:
Agent, Market, Resource allocation, Communications network, Scalability, Robustness
Organisations:
Web & Internet Science, Agents, Interactions & Complexity
Identifiers
Local EPrints ID: 261482
URI: http://eprints.soton.ac.uk/id/eprint/261482
ISBN: 2-913923-14-3
PURE UUID: 65b19e8c-850a-4a45-8d94-fd0a696b02d2
Catalogue record
Date deposited: 17 Oct 2005
Last modified: 14 Mar 2024 06:53
Export record
Contributors
Author:
Nadim Haque
Author:
N. R. Jennings
Author:
Luc Moreau
Editor:
Daniel Grosu
Editor:
Jonathan Shapiro
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