Build failed in Hudson: hbase-0.90 #11

See https://hudson.apache.org/hudson/job/hbase-0.90/11/changes Changes: [stack] HBASE-3298 Regionserver can close during a split causing double assignment -- fix broken heapsize test -- [...truncated 2734 lines...] Tests run: 2, Failures: 0, Errors: 0,

Build failed in Hudson: HBase-TRUNK #1681

See https://hudson.apache.org/hudson/job/HBase-TRUNK/1681/changes Changes: [jdcryans] HBASE-3309 Not running balancer because dead regionserver processing is a lie [stack] HBASE-3290 Regionserver can close during a split causing double assignment -- addendum [stack] Some fixup of hbck log

Local sockets

Hi, has anyone tested the performance impact (when there is a hdfs datanode and a hbase node on the same machine) of using unix domain sockets communication or shared memory ipc using nio? I guess this should make a difference on reads? Regards, Leen

Re: Local sockets

Hi Leen, Check out HDFS-347 for more info on this. I hope to pick this back up in 2011 - in 2010 we mostly focused on stability above performance in HBase's interactions with HDFS. Thanks -Todd On Sat, Dec 4, 2010 at 12:28 PM, Leen Toelen toe...@gmail.com wrote: Hi, has anyone tested the

Build failed in Hudson: HBase-TRUNK #1682

See https://hudson.apache.org/hudson/job/HBase-TRUNK/1682/changes Changes: [stack] HBASE-3298 Regionserver can close during a split causing double assignment -- fix broken heapsize test -- add reference too [stack] HBASE-3298 Regionserver can close during a split causing double assignment --

RE: Local sockets

From my own experiments performance difference is huge even on sequential R/W operations (up to 300%) when you do local File I/O vs HDFS File I/O Overhead of HDFS I/O is substantial to say the least. Best regards, Vladimir Rodionov Principal Platform Engineer Carrier IQ, www.carrieriq.com

Re: Local sockets

On Sat, Dec 4, 2010 at 2:57 PM, Vladimir Rodionov vrodio...@carrieriq.comwrote: From my own experiments performance difference is huge even on sequential R/W operations (up to 300%) when you do local File I/O vs HDFS File I/O Overhead of HDFS I/O is substantial to say the least. Much of

Re: Local sockets

While I applaud these experiments, the next challenge is getting them in to a shipping Hadoop. I think it's a relative nonstarter if we require someone to patch in a bunch of patches that are/were being refused to be committed. Keep on experimenting and collecting that evidence though! One day!

Build failed in Hudson: hbase-0.90 #12

See https://hudson.apache.org/hudson/job/hbase-0.90/12/changes Changes: [stack] HBASE-3298 Regionserver can close during a split causing double assignment -- fix broken heapsize test -- add reference too -- [...truncated 2734 lines...] Tests run: 2,