A (Simple) Big Data Architecture

How does a Big Data architecture looks like?

Of course, your requirements/tools surely can change, but this should give you a pretty good abstract view of the different layers applied in a Big Data environment. There are many things missing, like caching (memcached), data indexing (Solr), task execution (Zookeeper) and data extraction/analysis/mining software (there are so many).

You might choose not use a different NoSQL solution but directly use HBase for your NoSQL data or use different NoSQL software for different data (HBase & Neo4J).