Question

In my previous question I asked how to implement fixed number columns in cassandra with CQL.

The problem is, it doesn’t make sense that user tries to keep a row’s columns in a fixed amount(size).

We know that “wide row design” is very handy and on the other hand we have to keep size in about a reasonable size. Then, we need buckets right? Monthly, daily or even hourly buckets… The problem with buckets is the distribution that for some rows hourly for others may be yearly buckets make sense.

So why not to tell cassandra, I want a column family like LRU cache but on disk for some “most recent data”.

Column family storage parameters :

max_column_number_hint : 1000 // meaning: try to keep around 1000, will you? I am OK with tombstones or 800 – 1200 range

or

max_row_size_hint : 1MB

Cassandra has already background jobs and columns already has timestamps; so, again why not?

Answer 1

It’s there as CASSANDRA-3929 .

We currently support expiring columns by time-to-live; we’ve also had requests for keeping the most recent N columns in a row.