Home | About | Sematext search-lucene.com search-hadoop.com
clear query|facets|time Search criteria: .   Results from 131 to 140 from 374 (0.053s).
Loading phrases to help you
refine your search...
Re: Straggler map tasks - MapReduce - [mail # dev]
...Yes, you should look at JobInProgress class, and search for 'speculative'. JobTracker already has logic to launch speculative copies for tasks which are progressing at a degraded rate relati...
   Author: Vinod Kumar Vavilapalli, 2012-11-26, 18:40
Re: Job running on YARN gets automatically killed after 10-12 minutes - MapReduce - [mail # user]
...Is this your custom application and not, say, MapReduce or the distributed shell?  If that is the case, the ApplicationMaster needs to constantly ping the ResourceManager so that RM can...
   Author: Vinod Kumar Vavilapalli, 2012-11-05, 18:47
[MAPREDUCE-4761] MR AM doesn't handle exceptions from RM correctly - MapReduce - [issue]
...The MR AM currently ignores exceptions upto a certain count irrespective of the failure type.Exception handling on the MR AM side is totally broken, it doesn't seem to shutdown the AM on get...
http://issues.apache.org/jira/browse/MAPREDUCE-4761    Author: Vinod Kumar Vavilapalli, 2012-11-01, 00:39
Re: Shoud I use MapReduce 0.2X, or 1.0? - MapReduce - [mail # user]
...There are two major release lines: 1.x and 2.x.  1.x is deemed stable and 2.x is considered alpha quality.  0.23.* is the last of 0.2X line. 0.23.* is used mostly by Yahoo! to &nbs...
   Author: Vinod Kumar Vavilapalli, 2012-10-31, 19:11
Re: task trackers - MapReduce - [mail # user]
...Even JT accepts exclude/include hosts files for the slaves. If they  aren't populated, JT by default accepts TTs that successfully register  with it at runtime.  Thanks, +Vino...
   Author: Vinod Kumar Vavilapalli, 2012-10-22, 18:03
Re: Java heap space error - MapReduce - [mail # user]
...Did this job ever run successfully for you? With 200m heap size?  Seems like your maps are failing. Can you paste your settings for the  following:  - io.sort.factor  - i...
   Author: Vinod Kumar Vavilapalli, 2012-10-22, 18:01
Re: speculative execution in yarn - MapReduce - [mail # user]
...Speculative execution is a per-job concept, so in 2.* release line, it  is MR AM's responsibility.  Because there is a two level scheduling - one at RM and one at AM, AMs  hav...
   Author: Vinod Kumar Vavilapalli, 2012-10-17, 18:28
Re: GroupingComparator - MapReduce - [mail # user]
...On Oct 15, 2012, at 12:27 PM, Dave Beech wrote:    Yes, it is a bit counter intuitive. In the older API, there is no way to  find out the current key corresponding to the curr...
   Author: Vinod Kumar Vavilapalli, 2012-10-16, 18:44
Re: DFS respond very slow - MapReduce - [mail # user]
...I just realized one more thing. You mentioned "disk is 700Gb RAID".  How  many disks overall? What RAID configuration? Usually we advocate JBOD  with hadoop to avoid performan...
   Author: Vinod Kumar Vavilapalli, 2012-10-16, 02:36
Re: Suitability of HDFS for live file store - MapReduce - [mail # user]
...For your original use case, HDFS indeed sounded like an overkill. But  once you start thinking of thumbnail generation, PDFs etc, MapReduce  obviously fits the bill.  If you w...
   Author: Vinod Kumar Vavilapalli, 2012-10-15, 23:25
MapReduce (362)
YARN (133)
Hadoop (122)
Ambari (29)
HDFS (27)
Hive (17)
Bigtop (1)
Chukwa (1)
HBase (1)
issue (260)
mail # user (64)
mail # dev (50)
last 7 days (14)
last 30 days (33)
last 90 days (40)
last 6 months (48)
last 9 months (374)
Harsh J (454)
Arun C Murthy (326)
Vinod Kumar Vavilapalli (309)
Todd Lipcon (212)
Amar Kamat (181)
Thomas Graves (165)
Jason Lowe (157)
Amareshwari Sriramadasu (153)
Sandy Ryza (124)
Tom White (111)
Siddharth Seth (109)
Aaron Kimball (107)
Owen O'Malley (105)
Devaraj K (103)
Ramya Sunil (103)
Alejandro Abdelnur (101)
Robert Joseph Evans (101)
Hemanth Yamijala (97)
Steve Loughran (90)
Eli Collins (77)
Ravi Gummadi (76)
Ted Yu (76)
Karthik Kambatla (67)
Mahadev konar (67)
Ravi Prakash (66)