SQLS: A Storm-Based Query Language System for Real-Time Stream Data Analysis

In this work, a Storm-based query language System (SQLS) is proposed for real-time data stream analysis. The system is compatible with Continuous query language (CQL) specification. It supports both continuous queries and one-time queries over streaming data, and meets the requirements of user experience (traditional SQL queries) and QoS (such as real-time and throughput). In order to better meet the requirement of throughput and enhance the processing efficiency, the load shedding algorithm and cache optimization are employed during the implementation of SQL-like operators. Finally, performance testing of the proposed SQLS has been conducted on stand-alone Storm platform and Storm clusters. Experimental results show that our system can not only meet the needs of users, but also extend the function of real-time streaming queries processing.