- What are the recommendations for the configuration of Cassandra cluster?
- What are the pros and cons of the different possible configuration of Cassandra (for Dataminer)?
I understand from the help that Cassandra cluster is supported as the general database for a DataMiner System. Now, with DataMiner 10.1.0/10.1.2, we can have all DataMiner nodes in a cluster using the same Cassandra cluster as their general database.
https://help.dataminer.services/dataminer/index.html?#rhsyns=%20&t=DataMinerUserGuide%2Fpart_3%2Fdatabases%2FMigrating_the_general_database_to_a_DMS_Cassandra_cluster.htm&rhsearch=database%20per%20dma&rhhlterm=database%20per%20dma
Hi Yohann,
Following link provides an overview regarding the compute requirements for Cassandra (when running on the DMA server, or on separate nodes).
The advantage of running them on the same server as DataMiner is that you don't need additional servers. You will need additional resources on these servers since the Cassandra database puts additional load on the system. This is also the dissadvantage since in rare cases Cassandra can take resources that DataMiner needs.
The advantage of running the Cassandra nodes on other servers is that it can't influence the DataMiner setup as it's running on separate servers. You could therefore regard it as having less risk.
It's possible to run them on Linux servers. Cassandra is built for Linux servers and roll out of a Cassandra cluster is straight forward.
Dissadvantage is that you need additional server nodes.
What to recommend will largely depend on the size of the system and what the user will do with it.
Large systems with lots of data being generated will benefit from a separate Cassandra cluster.