Cassandra
Cassandra is a highly scalable, eventually consistent, distributed, structured row/column store.
Scylla
Scylla is a Cassandra-compatible NoSQL data store
Others
| Example |
|---|
|
The internal implementation of the memtables might be slightly different between scylla and cassandra but for the sake of simplicity let s assume it is the same from question Understanding SSTable immutiability |
|
Details here given your use case if you envision storing multiple tb of data i would recommend scylla over cassandra from question Database selection for Read-only and write-once(insert) every growing data not to be modified |
|
I wrote a detailed description of the sstable format on scylla s site scylla is a more efficient c++ re-implementation of cassandra to which i contribute from question Conceptual difference concerning column families in Cassandras data model compared to Bigtable? |
|
Scylla a cassandra clone which is generally more efficient than cassandra also has similar issues with huge partitions as in cassandra moderately large partitions are fine but similar issues are actively being worked on including re-designing the file format so eventually scylla a cassandra clone which is generally more efficient than cassandra should support arbitrary-sized partitions from question How to avoid slow query if data is too big for one partition in NoSQL? |
|
Scylla considered such an implementation but hasn t implemented it either see;there is a reason why cassandra doesn t support the feature you describe - it s because it can be inefficient first the secondary-index must indeed list as you noted all the matching row keys and not just distinct matching partition keys from question How to properly select partition keys that match an index |