Joel ask the following question:
=================
Hi All,
I have an application with two HBase tables.
One table is written to frequently, by a crawler writing web pages.
Another table is written to occasionally (the result of some
processing), but end users read data from this table, and I want the
read response times to be as low as possible.
I only have one server on which to host both tables.
What tuning should I consider to minimise the read latency on the second
table (there will be relatively few users, so throughput is less of a
concern, for the time being) ?
Regards,
Joel
Stack answer it at first
================
Mark the second-table in-memory in the schema. And for the first,
have it not use cache at all. This way, cache should only have
content from the table that is read. Does the second table fit fully
in memory?
Joel reply it again
================
No, the second table is too large to fit in memory.
Lars George answer it at second
================
Hi Joel,
Marking it "in-memory" is *not* making it all stay or be loaded into
memory. It is just a priority flag to retain blocks of that CF
preferably in the block caches. So it caches it up to the max block
cache size. The rest may cause some churn but that is the best you can
do.
Lars