Hey, You may already know about this: http://wiki.apache.org/hadoop/HBase/HBasePresentations
Many good things have been said in various presentations. Dont forget the list archives, lots of good, but unstructured data. The bigtable paper has tons of subtle hints, read it several times. Lots of dense but hidden data. I can give you pithy quotes: - optimize for your read pattern - dont use explicit row locks - more rows is better, since puts lock rows - more region is better to spread load across - use modern hardware and lots of ram -ryan On Fri, Aug 13, 2010 at 6:41 AM, Burak ISIKLI <[email protected]> wrote: > Thank you for your interest. But in fact i want to learn what should be data > model and how can i improve performance for that model. > > > > > > ________________________________ > From: Lekhnath <[email protected]> > To: "[email protected]" <[email protected]> > Sent: Fri, August 13, 2010 1:49:01 PM > Subject: Re: Hbase architecture > > On 8/13/2010 4:13 PM, Burak ISIKLI wrote: >> Hi; >> I want to learn the hbase background architecture. Can u suggest an article >> or >> something like that? >> thanks >> regards... >> >> >> >> > BigTable paper is a good starting document for the background architecture. > http://labs.google.com/papers/bigtable.html > For additional information regarding the details specific to HBase, you > can refer HBase wiki > http://wiki.apache.org/hadoop/Hbase/HbaseArchitecture > > good luck > > Lekhnath > > > > This email is intended for the recipient only. If you are not the intended > recipient please disregard, and do not use the information for any purpose. > > > >
