Comparative Analysis on Big Data Tools: Apache Solr Search and Hibernate Search

Abstract

In the following paper, an in-depth comparison of Apache Solr Search and Hibernate Search is presented. In the modern-day scenario, a lot of software enterprises require high performance and efficient searches in SQL based databases which consist of billions of entities. Since SQL's ‘like’ clause cannot provide these functionalities in the most efficient manner, several other libraries came into existence. Apache Lucene is one of the many common libraries used for the same in the software industry. It is a very powerful yet complex library with a wide variety of applications and advantages over conventional searches. Apache Solr Search and Hibernate Search are both implementations of Lucene with a variety of optimizations from the developers so as to facilitate its performance as per different requirements whilst making it user-friendly. In this paper, the search query execution times for both of the frameworks have been compared for several kinds of search methodologies on a test Big Data(110 Million Entries in MySQL) database. Upon comparison of the execution times, Solr Search was clearly performing better with Full-Text Searches, Fuzzy Searches, Wildcard Searches, etc. while Hibernate Search trumped when it came to Spatial Searches, Range Searches, etc. As both Solr and Hibernate Searches provide similar functionalities to some extent, software firms often struggle with picking a suitable one to best suit their needs. The following paper is directed towards helping such companies out by providing them with verified analysis and results on different types of searches, thereby helping them select the platform yielding most optimal results for them.