A Priori Study on Factors Affecting MapReduce Performance in Cloud-Based Environment

In the current era, global data have been rising at a very fast speed due to the excessive use of technologies including cloud and IoT. It leads to the development of big data that can handle and analyze a high volume of data regularly. Cloud computing provides a reliable, available, and scalable environment for the processing of this huge data. MapReduce has become an important computing model for processing and generating high-volume datasets on a cluster of machines. It not only allows distributed processing but also significant attributes like flexibility, versatility, load adjusting, and adaptation to internal failure. Despite these benefits, the performance of this framework gets affected by multiple factors, namely indexing, data skew, joining, caching, and load balancing. The main objective of this paper is to identify the factors to resolve its performance issues and investigate alternate strategies to improve the MapReduce query performance in the cloud-based environment.