OnCall: defeating spikes with a free-market application cluster Export

@inproceedings{citeulike:1802976, 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.}, author = {Norris, J. and Coleman, K. and Fox, A. and Candea, G.}, booktitle = {Autonomic Computing, 2004. Proceedings. International Conference on}, citeulike-article-id = {1802976}, citeulike-linkout-0 = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1301364}, journal = {Autonomic Computing, 2004. Proceedings. International Conference on}, keywords = {571g, allocation, based, market, resource}, pages = {198--205}, posted-at = {2007-10-21 23:41:02}, priority = {0}, title = {OnCall: defeating spikes with a free-market application cluster}, url = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1301364}, year = {2004} }

TY - CONF ID - citeulike:1802976 L3 - citeulike-article-id:1802976 N2 - 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. JF - Autonomic Computing, 2004. Proceedings. International Conference on EP - 205 TI - OnCall: defeating spikes with a free-market application cluster T2 - Autonomic Computing, 2004. Proceedings. International Conference on SP - 198 KW - 571g KW - allocation KW - based KW - market KW - resource AU - Norris, J AU - Coleman, K AU - Fox, A AU - Candea, G PY - 2004/// UR - http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1301364 ER -