IBM Big SQL, the best SQL on hadoop engine, now also supportsÂ HDP. With this, BigSQL is also ODPI runtime compliant. This supports provides greater simplicity to clients for their BigData solutions.
Â Big SQL is the IBM SQL interface in the Apache Hadoop environment. You can use Big SQL to summarize, query, and analyze data. Big SQL’s massively parallel processing (MPP) SQL engine deploys directly on the physical Hadoop Distributed File System (HDFS) cluster. Big SQL uses a low-latency parallel execution infrastructure that accesses Hadoop data natively for reading and writing. It provides full ANSI SQL 2013 support. Big SQL is the ultimate Platform for RDBMS off-load and consolidation. It’shybrid engine exploit’s Hive and all Hive storage formats. Big SQL with HBASE provides full INSERT, UPDATE, DELETE support.
BigSQL has the most sophisticated and mature query rewrite and cost based optimizer. It provides superior performance, especially in multi-user concurrent execution environments and supports SQL compliance with well known relational SQL engines including Oracle, Teradata, mySQL and Netezza.
IBM Big SQL is the first, and currently the only query engine that integrates with Spark. This allows users to blend sql with their non-relational data analytics. In Big SQL V4.2, Big SQL Head node(s) can start its own driver program and spawn Spark executors. The Spark jobs invocation from BigSQL â€“ the SYSHADOOP.EXECSPARK function can trigger the execution of a Spark job and use the results in a SQL query.
From Performance and Scalability ,in comparison to Hortonworks supported Hawq, Big SQL scales better as Data Volume increases for Multiple Users. The performance gap increases from 18% faster to 270% (2.7X) as volumes increase when queries are run concurrently.
Overall, IBM Big SQL is the best SQL on Hadoop engine for both IOP and HDP.