Re: [hbase] table HRegionServer affinity

2008-01-08 Thread Andrew Purtell
stack wrote: > You might be interested in the PNUTs description here: > http://wiki.apache.org/lucene-hadoop/Hbase/PNUTS. In > particular see section where it talks of how it does its > global distribution. > > St.Ack Thanks. Some interesting tidbits there, maybe. We are also looking at, as an

Re: [hbase] table HRegionServer affinity

2008-01-08 Thread stack
You might be interested in the PNUTs description here: http://wiki.apache.org/lucene-hadoop/Hbase/PNUTS. In particular see section where it talks of how it does its global distribution. St.Ack

RE: [hbase] table HRegionServer affinity

2008-01-08 Thread Andrew Purtell
Thank you all for your considered responses. I would like to make a few comments and clarifications (inline, below). Jim Kellerman wrote: > > Andrew Purtell wrote: > > First: How invasive to the HBase master/region model is > > the concept of specifying constraints on data mobility? > > It would

RE: [hbase] table HRegionServer affinity

2008-01-08 Thread Jim Kellerman
: hadoop-dev > Subject: [hbase] table HRegionServer affinity > > Hello, > > Consider the case of a global federation of Hadoop clusters, > with a single global HBase master, divided into a number of > geographic regions each with a local DFS, local workload, and > region serv

RE: [hbase] table HRegionServer affinity

2008-01-08 Thread edward yoon
-1 B. Regards, Edward yoon. > Date: Mon, 7 Jan 2008 20:49:29 -0800 > From: [EMAIL PROTECTED] > Subject: [hbase] table HRegionServer affinity > To: hadoop-dev@lucene.apache.org > > Hello, > > Consider the case of a global federation of Hadoop > clusters, with a

Re: [hbase] table HRegionServer affinity

2008-01-08 Thread Andrew Purtell
Billy Pearson wrote: > Would an option to fix a table to a set of regions work > in this case? > > > 1) The ability to fix a given key range to a region. > > This would both assign a range to a given region, and > > also disable splitting over that range. Aside from API > > changes, ideally there

Re: [hbase] table HRegionServer affinity

2008-01-07 Thread Billy
Would an option to fix a table to a set of regions work in this case? > 1) The ability to fix a given key range to a region. This > would both assign a range to a given region, and also > disable splitting over that range. Aside from API changes, > ideally there would be a HBase shell command to s

Re: [hbase] table HRegionServer affinity

2008-01-07 Thread stack
Andrew Purtell wrote: Going forward what would be the protocol for bug reports or discussions regarding perhaps minor enhancements? Would it be appropriate for me to submit an issue under contrib/hbase on JIRA? Thats the way to go (JIRAs with patches attached get more prompt attention; smi

Re: [hbase] table HRegionServer affinity

2008-01-07 Thread Andrew Purtell
> stack <[EMAIL PROTECTED]> wrote: > Andrew Purtell wrote: [...] > > 1) The ability to fix a given key range to a region. > > This would both assign a range to a given region, and > > also disable splitting over that range. Aside from API > > changes, ideally there would be a HBase shell command >

Re: [hbase] table HRegionServer affinity

2008-01-07 Thread stack
Andrew Purtell wrote: .. To be a bit more specific, I have three things in mind: 1) The ability to fix a given key range to a region. This would both assign a range to a given region, and also disable splitting over that range. Aside from API changes, ideally there would be a HBase shell comman

[hbase] table HRegionServer affinity

2008-01-07 Thread Andrew Purtell
Hello, Consider the case of a global federation of Hadoop clusters, with a single global HBase master, divided into a number of geographic regions each with a local DFS, local workload, and region server backed by that DFS. This setup allows for a global HBase space, where any region may retrieve