The case for Kudu from Cloudera

There is a gap between HDFS and HBase. This gap becomes obvious when we need to collect billions of events, analyze them and update/enrich them with information that arrives later on. The classic case is: adding to the click information about in-app purchase which arrived a week later.

Read only HDFS do not allow this. HBase does not provide fast analytics. And this is exactly the point where Kudu comes in.

In this Meetup we will have 2 full presentations and possibly another brief one. Those are:

1. Shlomi Tubul (from Cloudera) will explain about the place Kudu takes in the ecosystem and its use cases.

2. David Gruzman (from Nestlogic) will go deeper into Kudu itself - how it is built, its architecture, etc.

3. SimilarWeb (possible) will brief out their use of HBase for problems that are essentially designed to be solved with Kudu.