Thanks..we have the same exact code that process 700 million puts per day in 0.20.6 from a tomcat servlet (each thread creates new HTable, does 1 put & closes)
..in 0.90.2..we changed just the API whose signature changed (mainly HTable)..it crawls ..each/most requests taking well over 2 sec..we can't keep up with even 1/10 th of production load. everything in the cluster is identical 20 node cluster.. That is impressive performance..from async..Thanks for the tip..i'll give it a try (assuming it would work with 0.90.2) -----Original Message----- From: tsuna <tsuna...@gmail.com> To: user@hbase.apache.org Sent: Wed, Apr 20, 2011 4:30 pm Subject: Re: java.lang.IndexOutOfBoundsException On Wed, Apr 20, 2011 at 10:04 AM, Venkatesh <vramanatha...@aol.com> wrote: > On 0.90.2, do you all think using HTablePool would help with performance problem? What performance problems are you seeing? BTW, if you want a thread-safe client that's highly scalable for high-throughput, multi-threaded applications, look at asynchbase: http://github.com/stumbleupon/asynchbase OpenTSDB uses it and I'm able to push 200000 edits per second to 3 RegionServers. -- Benoit "tsuna" Sigoure Software Engineer @ www.StumbleUpon.com
