Cloud Computing and Resource Management

Cloud computing research examines how computing power, storage, and software can be delivered over networks from large shared facilities called data centers, and how the resources within those facilities — processors, memory, bandwidth — can be allocated efficiently across many simultaneous users and workloads. A central challenge is managing scale: frameworks like MapReduce and Hadoop were developed specifically to distribute the processing of massive datasets across thousands of machines in a coordinated way, while virtualization allows a single physical server to run many isolated software environments at once. Researchers are actively working on how to reduce the substantial energy consumption of data centers without sacrificing performance, and on how scheduling and resource allocation algorithms can respond dynamically to unpredictable demand. Open questions remain around making distributed systems more fault-tolerant, minimizing the latency that arises when data must travel between distant nodes, and fairly pricing shared infrastructure in ways that reflect actual resource use.

Works

132,806

Total citations

1,302,672

Keywords

Cloud ComputingBig DataMapReduceVirtualizationData CentersResource Management