Managing energy and server resources in hosting centers Export

@inproceedings{citeulike:2931455, abstract = {Internet hosting centers serve multiple service sites from a common hardware base. This paper presents the design and implementation of an architecture for resource management in a hosting center operating system, with an emphasis on energy as a driving resource management issue for large server clusters. The goals are to provision server resources for co-hosted services in a way that automatically adapts to offered load, improve the energy efficiency of server clusters by dynamically resizing the active server set, and respond to power supply disruptions or thermal events by degrading service in accordance with negotiated Service Level Agreements (SLAs).Our system is based on an economic approach to managing shared server resources, in which services "bid" for resources as a function of delivered performance. The system continuously monitors load and plans resource allotments by estimating the value of their effects on service performance. A greedy resource allocation algorithm adjusts resource prices to balance supply and demand, allocating resources to their most efficient use. A reconfigurable server switching infrastructure directs request traffic to the servers assigned to each service. Experimental results from a prototype confirm that the system adapts to offered load and resource availability, and can reduce server energy usage by 29\% or more for a typical Web workload.}, address = {New York, NY, USA}, author = {Chase, Jeffrey S. and Anderson, Darrell C. and Thakar, Prachi N. and Vahdat, Amin M. and Doyle, Ronald P.}, booktitle = {SOSP '01: Proceedings of the eighteenth ACM symposium on Operating systems principles}, citeulike-article-id = {2931455}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=502045}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/502034.502045}, comment = {Interesting use of technologies: * use of wake-on-lan to dynamically provision resources aside from allocating idle-power to machines. uses mac addresses to pinpoint which machine to turn-on * degrades QoS on negotiated service level agreements. this addresses the high cost of overprovisioning * use of economic models based on power consumption costs. tries to maximize profit based on throughput and latency as parameters * increased reliance on backup power to decrease power cost. * turning on a second machine to for load balancing purposes at a later time creates up to 37\% energy savings. Experiment used SURGE to generate bursty HTTP traffic for the web server.}, doi = {10.1145/502034.502045}, isbn = {1-58113-389-8}, keywords = {datacenter, energy, power}, location = {Banff, Alberta, Canada}, pages = {103--116}, posted-at = {2008-06-26 18:04:46}, priority = {2}, publisher = {ACM}, title = {Managing energy and server resources in hosting centers}, url = {http://dx.doi.org/10.1145/502034.502045}, year = {2001} }

TY - CONF ID - citeulike:2931455 L3 - citeulike-article-id:2931455 N2 - Internet hosting centers serve multiple service sites from a common hardware base. This paper presents the design and implementation of an architecture for resource management in a hosting center operating system, with an emphasis on energy as a driving resource management issue for large server clusters. The goals are to provision server resources for co-hosted services in a way that automatically adapts to offered load, improve the energy efficiency of server clusters by dynamically resizing the active server set, and respond to power supply disruptions or thermal events by degrading service in accordance with negotiated Service Level Agreements (SLAs).Our system is based on an economic approach to managing shared server resources, in which services "bid" for resources as a function of delivered performance. The system continuously monitors load and plans resource allotments by estimating the value of their effects on service performance. A greedy resource allocation algorithm adjusts resource prices to balance supply and demand, allocating resources to their most efficient use. A reconfigurable server switching infrastructure directs request traffic to the servers assigned to each service. Experimental results from a prototype confirm that the system adapts to offered load and resource availability, and can reduce server energy usage by 29% or more for a typical Web workload. CY - Banff, Alberta, Canada SN - 1-58113-389-8 AD - New York, NY, USA EP - 116 TI - Managing energy and server resources in hosting centers PB - ACM T2 - SOSP '01: Proceedings of the eighteenth ACM symposium on Operating systems principles SP - 103 KW - datacenter KW - energy KW - power N1 - Interesting use of technologies: * use of wake-on-lan to dynamically provision resources aside from allocating idle-power to machines. uses mac addresses to pinpoint which machine to turn-on * degrades QoS on negotiated service level agreements. this addresses the high cost of overprovisioning * use of economic models based on power consumption costs. tries to maximize profit based on throughput and latency as parameters * increased reliance on backup power to decrease power cost. * turning on a second machine to for load balancing purposes at a later time creates up to 37% energy savings. Experiment used SURGE to generate bursty HTTP traffic for the web server. AU - Chase, Jeffrey AU - Anderson, Darrell AU - Thakar, Prachi AU - Vahdat, Amin AU - Doyle, Ronald PY - 2001/// UR - http://dx.doi.org/10.1145/502034.502045 ER -