On the energy (in)efficiency of Hadoop clusters

382 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Quasar: resource-efficient and QoS-aware cluster management (2014, background): “Underutilized servers contribute to capital expenses and, since they are not energy proportional [36, 42], to operational expenses as well.”
Toward Scalable Systems for Big Data Analytics: A Technology Tutorial (2014, background): “The Covering-Set approach [274] designates certain nodes to host at least a replica of each data block, and other nodes are powered off during low-utilization periods.”
GreenHadoop: leveraging green energy in data-processing frameworks (2012, methodology): “We compare GreenHadoop’s data management against the covering subset approach [20].”
Parasol and GreenSwitch: managing datacenters powered by renewable energy (2013, methodology): “We configure our Hadoop setup to the Covering Subset scheme [27], i.”
The Case for Evaluating MapReduce Performance Using Workload Suites (2011, background): “Studies on MapReduce energy efficiency [22], [23] can quantify energy savings under realistic workload fluctuations.”
Towards energy proportionality for large-scale latency-critical workloads (2014, background): “Managing MapReduce clusters for energy efficiency has also been widely explored[24, 22, 9], but these consolidationbased techniques are not applicable for OLDI workloads with tight latency constraints and large dataset footprints.”