ElasticTree.pptVIP

ElasticTree: Saving Energy in Data Center Networks Introduction Most efforts to reduce energy consumption in Data Centers is focused on servers and cooling, which account for about 70% of a data center’s total power budget. This paper focuses on reducing network power consumption, which consumes 10-20% of the total power. 3 billion kWh in 2006 Data Center Networks There’s potential for power savings in data center networks due to two main reasons: Networks are over provisioned for worst case load Newer network topologies Over Provisioning Data centers are typically provisioned for peak workload, and run well below capacity most of the time. Rare events may cause traffic to hit the peak capacity, but most of the time traffic can be satisfied by a subset of the network links and switches. Network Topologies The price difference between commodity and non-commodity switches provides strong incentive to build large scale communication networks from many small commodity switches, rather than fewer larger and more expensive ones. With an increase in the number of switches and links, there are more opportunities for shutting down network elements. Typical Data Center Network Fat-Tree Topology Energy Proportionality Today’s network elements are not energy proportional Fixed overheads such as fans, switch chips, and transceivers waste power at low loads. Approach: a network of on-off non-proportional elements can act as an energy proportional ensemble. Turn off the links and switches that we don’t need to keep available only as much capacity as required. ElasticTree Example Optimizers The authors developed three different methods for computing a minimum-power network subset: Formal Model Greedy-Bin Packing Topology-aware Heuristic Formal Model Extension of the standard multi-commodity flow (MCF) problem with additional constraints which force flows to be assigned to only active links and switches. Objective function: Formal Model MCF problem is NP-complete An instance of the