Policy-based autonomous bidding for overload management in eCommerce websites
Author
Moreno, A.; Poggi, N.; Berral, J.; Gavaldà, R.; Torres, J.
Type of activity
Presentation of work at congresses
Name of edition
Group Decision and Negotiation Meeting
Date of publication
2007
Book of congress proceedings
Group decision and negotiation 2007: proceedings
First page
162
Last page
166
Abstract
In eCommerce applications with heterogeneous traffic, which typically run on execution environments where resources are shared with other applications, being able to dynamically adapt to the application workload in real time is crucial for proper self-management and business efficiency. The AUGURES platform has recently introduced a novel approach to deal with server overload situations, based on the prioritization of sessions according to the expected revenue that the session is likely to gener...
In eCommerce applications with heterogeneous traffic, which typically run on execution environments where resources are shared with other applications, being able to dynamically adapt to the application workload in real time is crucial for proper self-management and business efficiency. The AUGURES platform has recently introduced a novel approach to deal with server overload situations, based on the prioritization of sessions according to the expected revenue that the session is likely to generate. However, by denying access to exceeding low priority users, the website may be loosing potential customers, while there might be other resources available in the market of the execution environment, be it a server, a server farm, or a grid. This paper presents an extension of the AUGURES architecture with a simple policy-based autonomous bidding agent that generates automated bids for the extra resources needed to execute the transactions that have been refused by the server due to overload.