Analysing Performance in Retrieving Heterogeneous Information in Big Data Environment

Background/Objectives: The development of various communication media has generated few problems in retrieving information. The objective of the study is to analyze the performance of retrieving heterogeneous data. Methods/Statistical Analysis: A model was run to simulate the process of retrieving heterogeneous data from several servers. The information was distributed with different load and the servers were randomly selected. The performance had been analysed based on response time and CPU utilisation. A few types of load balancing techniques were applied to distribute the loads among the servers. The impacts on the overall system performance were discussed. Findings: Retrieving data requires high speed, where the response time must be very fast. The performance of retrieving heterogeneous data is a challenge, when servers have high load. When the load balancing techniques were not applied, some of the servers handle the entire load and the other servers have not been fully utilised. The results showed the response time decrease drastically when high load of data were applied to the server. When the load balancing was applied, the results were compared and presented. The results showed an improvement in the overall performance. Improvements/Applications: The load balancing techniques were applied based on several approaches. It allows an improvement in distributing the server load, which results in improvement in the performance.