OPTIMIZING ENERGY CONSUMPTION OF CLOUD COMPUTING SYSTEMS

A large number of network devices in cloud data-centers consume huge amount of energy. This has emerged as a matter of concern for data center operators. Cloud computing finds its application in various elds such as telecommunication, multimedia, video gaming and etc. Most of these applications require a timing guarantee which are reffed to as real-time applications. Once cloud providers aim to guarantee Quality of Service (QoS) parameters such as deadlines, more hard-ware resources are required because of using a higher number of servers, switches, higher network bandwidth, more cooling and power distribution systems. As a consequence of the use of a higher number of devices and communication links, power consumption and operational cost. This thesis aims to optimize the network energy consumption, which in turn results in reduction of the energy consumption of a data-center while respecting the network constraints. The energy reduction of the network is achieved by turning off idle network devices and playing with the allocation of the workload among the servers to minimize the communication load. Thereby, it increases the number of idle devices. Our experimental results demonstrate a full performance by using Open Shortest Path First (OSPF) routing protocol. OSPF creates a routing table to nd the nest tasks paths and the address of the idle devices. It provides 22.75% reduction in energy consumption by applying Elastic Tree mechanism. The performance of the method is also examined in terms of the network load. The results reveal a trade-off between network energy consumption and the load of the network. The results show that although the Elastic Tree mechanism optimizes energy consumption of the network, it increases the network load