Apache Phoenix

What do you like best?

SQL like syntax is the only thing positive

What do you dislike?

Apache Phoenix is still very primitive and does not offer easy SQL syntaxes. The errors/ exceptions are hard to understand, and development is slow.

Recommendations to others considering the product:

Do not expect much support from the community. Have created a ticket for over one and a half years and still it is unattended. Would have been great for the ticket to be solved as it would have solved a very common problem in big data loading into HBase tables via Apache Phoenix's Hadoop Map-Reduce program. Have provided detailed screenshots, steps to reproduce, etc. But when checked a few days ago, the ticket has not been picked up. Hence, I think it will be very painful for anyone to do advanced stuff with Apache Phoenix in the long run.

What problems are you solving with the product? What benefits have you realized?

Was trying to process our OLTP big data store. The only benefits are the SQL-like syntaxes and the that it can talk to HBase easily. The most positive point is that HBase and Apache Phoenix go hand in hand and so it is fairly easy to integrate your HBase tables with Apache Phoenix. The upsert statements are also helpful and there is a handly Hadoop Map Reduce program that you can execute from your Java program to load gigabytes of data to your HBase tables. This Map Reduce feature is the most beneficial aspect I have found off of Apache Phoenix. Note that, you can also use the command line to do the same operation.