The question
I am currently designing a data architecture for a real-time analytics project and I am confused about when to use HDFS versus when to implement HBase. My understanding is that both are part of the Hadoop ecosystem, but I don't quite grasp the functional distinction. Is HDFS just the storage layer while HBase is the database that runs on top of it? Specifically, how do they differ in terms of data latency, read/write patterns, and support for random access versus batch processing?
