Can I do [Oracle Hash Cluster] in Cassandra with comparable performance? E.g. One logical read to get the data by row key. about 100 millions rows, 2GB in size, so we can put all data in Cache (MemTable).
We run 2 millions single row INSERT or DELETE on peak hours, that slows the read response time in Oracle database, causes contention on Buffer busy wait.
We don't need real time consistency, a few minutes lag is acceptable.
Here is the use case on our Oracle database, we run the example query 200 times per second.
CREATE CLUSTER BOOK_CLUS
CREATE TABLE BOOK_CHECK
where book_id in (:1, :2, :3, :4, ... :30, :31);