DocumentCode :
3019234
Title :
OnCall: defeating spikes with a free-market application cluster
Author :
Norris, James ; Coleman, Keith ; Fox, Armando ; Candea, George
Author_Institution :
Dept. of Comput. Sci., Stanford Univ., CA, USA
fYear :
2004
fDate :
17-18 May 2004
Firstpage :
198
Lastpage :
205
Abstract :
Even with reasonable overprovisioning, today\´s Internet application clusters are unable to handle major traffic spikes and flash crowds. As an alternative to fixed-size, dedicated clusters, we propose a dynamically-shared application cluster model based on virtual machines. The system is dubbed "OnCall" for the extra computing capacity that is always on call in case of traffic spikes. OnCall\´s approach to spike management relies on the use of an economically-efficient marketplace of cluster resources. OnCall works autonomically by allowing applications to trade computing capacity on a free market through the use of automated market policies; the appropriate applications are then automatically activated on the traded nodes. As demonstrated in our prototype implementation, OnCall allows applications to handle spikes while still maintaining inter-application performance isolation and providing useful resource guarantees to all applications on the cluster.
Keywords :
Internet; computer network management; telecommunication traffic; virtual machines; workstation clusters; Internet clusters; OnCall; automated market policy; cluster resources; dedicated clusters; flash crowds; free-market application cluster; spike management; traffic spikes; virtual machines; Application software; Computer crime; Computer science; Internet; Resource management; Steady-state; Surges; Traffic control; Virtual machining; Voice mail;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Autonomic Computing, 2004. Proceedings. International Conference on
Print_ISBN :
0-7695-2114-2
Type :
conf
DOI :
10.1109/ICAC.2004.1301364
Filename :
1301364
Link To Document :
بازگشت