Hi Ajay, You won't be able to get unbiased opinion here easily. You'll need to try and see how each works for your use case. We use HBase for the SPM backend and it has worked well for us - it's stable, handles billions and billions of rows (I lost track of the actual number many moons ago) and fast, if you get your key design right. For our particular use case, HBase turned out to be a better choice, but we looked at Cassandra, too, back when we chose HBase. I'll answer your Q about monitoring:
I'd say both are equally well "monitorable". SPM <http://sematext.com/spm> can monitor both HBase and Cassandra equally well. Because Cassandra is a bit simpler (vs. HBase having multiple processes one needs to run), it's a bit simpler to add monitoring to Cassandra, but the difference is small. SPM is at http://sematext.com/spm if you want to have a look. We expose our own HBase clusters in the live demo, so you can see what metrics HBase exposes. We don't run Cassandra, so we can't show its graphs, but you can see some charts, metrics, and filters for Cassandra at http://blog.sematext.com/2014/06/02/announcement-cassandra-performance-monitoring-in-spm/ I hope this helps. Otis -- Monitoring * Alerting * Anomaly Detection * Centralized Log Management Solr & Elasticsearch Support * http://sematext.com/ On Fri, May 29, 2015 at 3:09 PM, Ajay <ajay.ga...@gmail.com> wrote: > Hi, > > I need some info on Hbase vs Cassandra as a data store (in general plus > specific to time series data). > > The comparison in the following helps: > 1: features > 2: deployment and monitoring > 3: performance > 4: anything else > > Thanks > Ajay >
