• Suyash Mishra
• Dr Suryakant yadav

The volume of data produced and used in today’s world has been growing very rapidly which has forced industry to explore data for better business decision making and incase profit by focusing more on customer needs. MapReduce is the core-processing engine of Hadoop, which accommodate rapidly increasing demands on computing resources required by massive data sets. Highly scalability of the MapReduce processing, allows parallel and distributed processing on multiple computing. Normally Hadoop implementation considers that underline cluster nodes are same in computing ability, configurations and storage. However, this homogeneity of cluster is not necessary every case and performance of MapReduce degrades and suffers due to various limitations in heterogeneous environments where underline nodes have different capability. This paper talks about how in heterogeneous environment, LATE (Longest Approximate Time to End) scheduling performs better and efficiently in comparison to other scheduling .LATE can improve Hadoop response times by almost two times in a clusters.

"Improving MapReduce Performance using LATE scheduling in big data.doc", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.360-364, June 2019, Available :http://www.jetir.org/papers/JETIR1908799.pdf

"Improving MapReduce Performance using LATE scheduling in big data.doc", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp360-364, June 2019, Available at : http://www.jetir.org/papers/JETIR1908799.pdf