If there're (many) other tables besides table A, the data may not be evenly
distributed across cluster.
See https://issues.apache.org/jira/browse/HBASE-3373
On Sat, Jan 1, 2011 at 2:46 AM, Eric wrote:
> I have little experience with HBase so far, but my feeling says it should
> not matter how m
I have little experience with HBase so far, but my feeling says it should
not matter how much rows you store and that it's better to save on cpu time
and bandwidth. HBase will distribute the data evenly over your cluster and
should be very good at making rows accessible quickly by key because it's
Hi all,
I have a schema design question that I would like to run by the group.
Table "A" has a composite row key containing a user identifier and an item
identifier.
I need to query this table by time, to determine the latest X new or updated
rows (on a per-user basis). (This is a "live" query
