Home | About | Sematext search-lucene.com search-hadoop.com
clear query|facets|time Search criteria: .   Results from 141 to 150 from 197 (0.078s).
Loading phrases to help you
refine your search...
Re: NN memory consumption on 0.20/0.21 with compressed pointers/ - Hadoop - [mail # user]
...On Tue, Aug 25, 2009 at 2:57 AM, Steve Loughran  wrote:    Since the codes running on these machines tend to be matrix oriented Fortran, I would expect almost all of this is a...
   Author: Ted Dunning, 2009-08-25, 15:59
Re: File Chunk to Map Thread Association - Hadoop - [mail # user]
...Uhh.... hadoop already goes to considerable lengths to make sure that computation is local.  In my experience it is common for 90% of the map invocations to be working from local data. ...
   Author: Ted Dunning, 2009-08-20, 17:36
Re: How to deal with "too many fetch failures"? - Hadoop - [mail # user]
...I think that the problem that I am remembering was due to poor recovery from this problem.  The underlying fault is likely due to poor connectivity between your machines.  Test tha...
   Author: Ted Dunning, 2009-08-20, 06:25
Re: Ubuntu/Hadoop incompatibilities? - Hadoop - [mail # user]
...I use ubuntu both in-house and on EC2 for hadoop.  Zero problems once you have the real java.  On Mon, Aug 17, 2009 at 10:34 AM, Jakob Homan  wrote:  ...
   Author: Ted Dunning, 2009-08-17, 17:46
Re: Why the jobs are suspended when I add new nodes? - Hadoop - [mail # user]
...Have you looked at the logs?  On Sun, Aug 16, 2009 at 11:36 PM, yang song  wrote:  ...
   Author: Ted Dunning, 2009-08-17, 08:00
Re: How to re-read the config files - Hadoop - [mail # user]
...You can do a rolling restart of the nodes.  The customer won't notice and running programs will still complete in good order.  If you have rack awareness configured, you can restar...
   Author: Ted Dunning, 2009-08-13, 22:21
Re: What will we encounter if we add a lot of nodes into the current cluster? - Hadoop - [mail # user]
...There is a parameter (dfs.balance.bandwidthPerSec) that limits the rebalancing bandwidth.  The default is rather low.  See http://developer.yahoo.com/hadoop/tutorial/module2.html#r...
   Author: Ted Dunning, 2009-08-13, 05:25
Re: How to break a hadoop-cluster in subclusters (how to group physical nodes)? - Hadoop - [mail # user]
...On Sun, Aug 9, 2009 at 8:17 AM, Harold Valdivia Garcia  wrote:    I think so.  In this configuration as you say I'd loss data-locatily because map-task   Yes.  ...
   Author: Ted Dunning, 2009-08-09, 18:37
Re: How to redistribute files on HDFS after adding new machines to cluster? - Hadoop - [mail # user]
...I think that I remember that you essentially doubled your storage before starting balancing.  This means that about 1 TB will need to be copied.  By default the balancer only moves...
   Author: Ted Dunning, 2009-08-08, 05:42
Re: Counting no. of keys. - Hadoop - [mail # user]
...If you need the number of token occurrences, then counters work well.  If you need the number unique tokens, then you need a separate map reduce.   On Mon, Aug 3, 2009 at 6:01 AM, ...
   Author: Ted Dunning, 2009-08-04, 14:25
Drill (246)
Zookeeper (242)
Hadoop (192)
HBase (134)
Pig (38)
HDFS (37)
MapReduce (35)
Chukwa (1)
mail # user (136)
mail # general (33)
mail # dev (27)
issue (1)
last 7 days (0)
last 30 days (0)
last 90 days (0)
last 6 months (0)
last 9 months (197)
Harsh J (537)
Owen O'Malley (402)
Steve Loughran (358)
Todd Lipcon (234)
Eli Collins (181)
Arun C Murthy (157)
Chris Nauroth (129)
Alejandro Abdelnur (121)
Allen Wittenauer (115)
Nigel Daley (112)
Tom White (111)
Daryn Sharp (108)
Konstantin Shvachko (102)
Ted Yu (95)
Aaron Kimball (93)
Ted Dunning