cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Robert Stupp (JIRA)" <j...@apache.org>
Subject [jira] [Created] (CASSANDRA-9336) Partition paging
Date Fri, 08 May 2015 21:39:00 GMT
Robert Stupp created CASSANDRA-9336:
---------------------------------------

             Summary: Partition paging
                 Key: CASSANDRA-9336
                 URL: https://issues.apache.org/jira/browse/CASSANDRA-9336
             Project: Cassandra
          Issue Type: New Feature
            Reporter: Robert Stupp
            Priority: Minor


A common problem in data modeling is how to prevent too-big partitions. For example, a table
usually has 100 rows per partition, but outliers have millions of rows per partition. One
"generic" way could be to split the outliers into separate buckets - but reading these outliers
may involve multiple read requests (which is bad).

CASSANDRA-9231 could provide some base functionality - i.e. allow users to "bind" different
partitions to the same nodes.

Making it possible to select all or a range of partitions per routing-key could help - ideally
with sorting. (Comparable to what a clustering key is for a partition.)



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message