Cassandra column family bigger than nodes drive space -
i want start cassandra cluster (eg. 3 nodes), , application has 1 column family.
after reading documentation, understood how cassandra replicates 1 column family across multiple nodes.
for eg. every node has 2tb drive space , column family replicated on every node every node contains full copy of it.
what happens if after years column family exceed 2tb , not have posibility increase drive space?
if add more 10 nodes, want column family splitted parts , stored on different drives on nodes, can increase infinite size. if understood correctly, column family limited smallest drive space in cluster?
the scenario describe case when data replicated nodes. configure setting replication factor (rf) number of nodes.
however, rf can less number of nodes , not need scale if add more nodes.
for example, if today had 3 nodes rf 3, each node contain copy of data, say. if add 3 more nodes , keep rf @ 3, each node have half data. can keep adding more nodes each node contains smaller , smaller proportion of data.
therefore there no limit in principle how big data can be.
Comments
Post a Comment