NOSQL Workload Analysis on Nosql Document Store Database Implementation Using Orientdb

Workload Analysis on Nosql Document Store Database Implementation Using Orientdb

Abstract

The development of the digitalization world currently has an impact on the data generated more and more from time to time so that the phenomenon of big data appears. Big Data is a collection of very large data that allows the processing, storage, and analysis of the data requires a very long time. This makes the RDBMS infrastructure that has been a mainstay in data storage and management has limitations in dealing with big data, so we need a new technology known as NoSQL to handle big data. One of the databases that handles NoSQL data storage and management is OrientDB. In this final project, a performance test of infrastructure built using YCSB's Core Workload will be tested, with variations in data size of (500,000, 1,000,000, 1,500,000 and 2,000,000 records). Each YCSB workload core consists of workloads A, B, C, D, E, F, G and H. The test is done by executing the load process that is insert data and run consisting of read, insert, read-modify-write, and update. Based on testing that has been done, single-node workload performance testing has better performance than multi-node performance. The addition of nodes in each workload test results in longer testing, this is due to the increasing number of nodes, the workload of the device becomes larger and requires a long time in processing YCSB. Keywords: Big Data, NoSQL, OrientDB, YCSB.