Energy Efficient Failure Recovery with Concurrent Execution in Hadoop Cluster Das Subrata Kumar1,*, Roy Saumendu2,**, Sarker Debbroto3,***, Rahman Md. Mijanur1,**** 1Department of Computer Science and Engineering, Jatiya Kabi Kazi Nazrul Islam University, Trishal-2220, Mymensingh, Bangladesh 2Department of Computer Science and Engineering, Jessore University of Science and Technology, Jessore-7408, Khulna, Bangladesh 3Department of Computer Science and Engineering, Jessore University of Science and Technology, Jessore-7408, Khulna, Bangladesh *(*Corresponding author) Email id: sdas_ce@yahoo.com
**saumocse3j6@gmail.com
***debu.dewdrop@gmail.com
****mijanjkkniu@gmail.com
Abstract This paper is mainly focused on the energy efficient cluster-based data processing. Since day-by-day a huge number of data are being placed in servers worldwide, there is a need of extra data processing. But for using commodity hardware for slave nodes, there is enough possibility of server node failure in hadoop cluster. Though different ways of data recovery are there, the need of energy for server node is another matter of concern. So, energy efficient failure recovery procedure has been introduced in which energy efficiency can be achieved by increasing workload into the server processor with concurrent execution process. Moreover, in this cluster, covering and non-covering subset-based platform is used. In this technique, the waiting time of covering subset used for recovering data is reduced besides power consumption. Top Keywords FS (fundamental set), ES (extended set), WS (waiting set), HDFS (hadoop distributed file system), Mapreduce. Top |