What about launching a map/reduce task from the thrift API?
For example, launch RowCount and either get the results sync in the
response or async from a results table.
Would this be a good way of handling load distribution of queries?
On Aug 4, 2008, at 9:50 AM, Andrew Purtell wrote:
Something to think about is integration of Thrift with the master
and regionserver themselves as a first class API. I think the
Thrift (and also the REST) APIs as clients/front ends are proof-
of-concepts more than anything else.
- Andy
From: Jean-Daniel Cryans <[EMAIL PROTECTED]>
Subject: Re: Question about how queries are distributed
To: [email protected]
Date: Monday, August 4, 2008, 6:05 AM
Ah ok I see what you meant! Yes, the Thrift client
communicates with a Thrift server which is bundled with
the Master, so the HBase client code doesn't run on your
local machine that queries HBase.
So yes, there may be a scalability problem
