[cygnus-ngsi][OrionCassandraSink] Any plan to support Apache Cassandra in the future?

[LeonanCarvalho]

Hi Everyone,
Apache Cassandra is a database engine with high availability,scalability and performance. There is a lot of production deployments, one of the largest production deployments is Apple's, with over 75,000 nodes storing over 10 PB of data. Other large Cassandra installations include Netflix (2,500 nodes, 420 TB, over 1 trillion requests per day), Chinese search engine Easou (270 nodes, 300 TB, over 800 million requests per day), and eBay (over 100 nodes, 250 TB). According to information from your website http://cassandra.apache.org/

Like MySQL, Hadoop and MongoDB, I think this DB enginer should be nice to have it on cygnus.

Best Regards.

[frbattid]

Hi Leonan, how are you doing? :)

We'll think on your suggestion. Of course, if you are interested on contributing, you are welcome :)

[cesarjorgemartinez]

+1

[DerJayDee]

+1

[DerJayDee]

Is the AGPL compatible with Apache 2.0? If so a Cassandra Sink might be easily implimented by using the following JDBC: http://mvnrepository.com/artifact/org.apache-extras.cassandra-jdbc/cassandra-jdbc

[LeonanCarvalho]

I'm out of the source code and architecture of Cygnus, but I can contribute in some way, sure!

[frbattid]

Having such a JDBC connector, developing a new OrionCassandraSink will be easy, mainly inspired in current existent OrionMySQLSink (or OrionPostgreSQLSink): instead of databases, we have keyspaces, then tables are created as well.

This kind of sink is not within our priorities, but I will be glad to review someone's contribution.

[DerJayDee]

I created a pull request ( [https://github.com//pull/899](Cassandra Pull) ) where i adapted the PostgreSQL sink to work with a Cassandra JDBC.

It is currently not tested in a real environment. If you could give me a hint on how to build up and install that Cygnus version I would gladly test it.

[frbattid]

Self assigning just for reviewing purposes.

[LeonanCarvalho]

Dear,
@DerJayDee, @frbattid how it going on?

Now, I can give more strength to this question, we realized that Cassandra is the best option for what we are proposing in our project about Government and City as a Platform.

The DerJayDee's pull request is conflicting of recent version, so I'm not able to build it or develop the integration usage.

I would like to build and/or deploy a Cygnus with CassandraSink and make some tests with our data points historical data.

[frbattid]

Dear Leonan, the PR is conflicting because the base branch used for developing the code, develop, has evolved a little bit in the past months... :) An important refactor was done, which involved a refactoring of the directory structure. Anyway, this should be something easy to fix by @DerJayDee:

• Now, everything related to NGSI persistence is under cygnus-ngsi.
• The documentation must be under doc/cygnus-ngsi.

BTW, sorry for the delay in reviewing the PR :( A lot of stuff had to be done during this time.