Improving the Performance of SQL Join Operation in the Distributed Enterprise Information System by Caching

The enterprise information system (EIS) contains databases and other data sources in multiple data centers. Users query the EIS via clients. The client has a working space in the cloud. Caching data in client space will reduce the total execution time of the query. However, the client space has limited resources to store data. There are two options for caching data at the client space: caching the final results of query operations, or caching the source data tables. The problem is that some query operations such as “joining multiple big tables” will simply produce a result too big to store in cache in some cases. By contrast, caching source data tables may be a better choice in those situations. This paper presents an algorithm that combines active caching and passive caching to improve the cache hit, thus improving performance of the SQL join query in the cloud computing environment.