Apache Trafodion is providing support for transactional or operational workloads on Hadoop!
What do you like best?
The best part is, it is a open source project by HP. and It supports ANSI SQL language and It comes with several component-specific testing libraries.
Second, Just Begin, commit, and rollback work syntax, including SET TRANSACTION.
Third, Multiple SQL processes participating in the same transaction concurrently.
Fourth, JDBC/ODBC connectivity with Linux/Windows OS.
What do you dislike?
Main cons is that it is very difficult to read logs for new comers. but hopefully it will be resolved soon in the next version.
Second, If you have not run any regression suites so far, then you will not have the required sub directories set up. you have to manually create them for each suite you want to run.
Recommendations to others considering the product:
Apache Trafodion builds on the scalability, elasticity, and flexibility of Hadoop. Apache Trafodion extends Hadoop to provide guaranteed transactional integrity, enabling new kinds of big data applications to run on Hadoop.
As it is a open source, enterprise class SQL-On-Hbase, so I will recommend to others to give it a trail run. It runs on a full-functional ANSI SQL language so it will defenitly work better in aspect of the others applications.
What problems are you solving with the product? What benefits have you realized?
To control transactional integrity and operational workloads for new kinds of Big Data applications that run on Hadoop.