Learning Application Models for Utility Resource Planning

@inproceedings{ShBaCh2006-LearningAppModels4UtilityResourcePlanning, abstract = {Shared computing utilities allocate compute, network and storage resources to competing applications on demand. An awareness of the demands and behaviors of the hosted applications can help the system to manage its resources more effectively. This paper proposes an active learning approach that analyzes performance histories to build predictive models of frequently used applications; the histories consist of measures gathered from noninvasive instrumentation on previous runs with varying assignments of compute, network and storage resources. An initial prototype uses linear regression to predict application interactions with candidate resources and combines them to forecast completion time for a candidate resource assignment. Experimental results from the prototype show that the mean forecasting errors range from 1\% to 11\% for a set of batch tasks captured from a production cluster. Examples illustrate how a system can use the learned models to guide task placement and data staging.}, author = {Shivam, P. and Babu, S. and Chase, J. S. }, booktitle = {ICAC '06: Proceedings of the IEEE International Conference on Autonomic Computing}, citeulike-article-id = {1842823}, journal = {Autonomic Computing, 2006. ICAC '06. IEEE International Conference on}, keywords = {autonomic-computing, resource-management, self-managed-systems, utility}, pages = {255--264}, posted-at = {2007-10-30 17:07:59}, priority = {2}, title = {Learning Application Models for Utility Resource Planning}, url = {http://ieeexplore.ieee.org/xpls/abs\_all.jsp?arnumber=1662406}, year = {2006} }

TY - CONF ID - ShBaCh2006-LearningAppModels4UtilityResourcePlanning L3 - citeulike-article-id:1842823 TI - Learning Application Models for Utility Resource Planning T2 - ICAC '06: Proceedings of the IEEE International Conference on Autonomic Computing JF - Autonomic Computing, 2006. ICAC '06. IEEE International Conference on SP - 255 EP - 264 N2 - Shared computing utilities allocate compute, network and storage resources to competing applications on demand. An awareness of the demands and behaviors of the hosted applications can help the system to manage its resources more effectively. This paper proposes an active learning approach that analyzes performance histories to build predictive models of frequently used applications; the histories consist of measures gathered from noninvasive instrumentation on previous runs with varying assignments of compute, network and storage resources. An initial prototype uses linear regression to predict application interactions with candidate resources and combines them to forecast completion time for a candidate resource assignment. Experimental results from the prototype show that the mean forecasting errors range from 1% to 11% for a set of batch tasks captured from a production cluster. Examples illustrate how a system can use the learned models to guide task placement and data staging. KW - autonomic-computing KW - resource-management KW - self-managed-systems KW - utility AU - Shivam, P AU - Babu, S AU - Chase, JS PY - 2006/// UR - http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1662406 ER -