Home | About | Sematext search-lucene.com search-hadoop.com
clear query|facets|time Search criteria: .   Results from 101 to 110 from 129 (0.173s).
Loading phrases to help you
refine your search...
RE: Shared HDFS for HBase and MapReduce - HBase - [mail # dev]
...Sure,  limiting number of slots is a way of IO throttling for MR jobs If you can do this - go ahead and do this.  Best regards, Vladimir Rodionov Principal Platform Engineer Carrie...
   Author: Vladimir Rodionov, 2012-06-06, 17:49
RE: Coprocessor - HBase - [mail # dev]
...I think the combination of a Scanner on a client side  and Coprocessor (Reg ionObserver) on HBase side will provide streaming functionality you need.  Your coprocessor can intercep...
   Author: Vladimir Rodionov, 2012-06-03, 18:24
RE: HBase backup option - HBase - [mail # dev]
...Carlos,  By "consistency" I meant - internal HBase consistency between .META. table  and table's WAL. I realize that w/o internal HBase & HDFS support for atomic snapshots there is...
   Author: Vladimir Rodionov, 2012-05-09, 18:03
RE: keyvalue cache - HBase - [mail # dev]
..."in the mean time, memcached could provide all those benefits without addin g any complexity to hbase..."  Tell this to Hypertable guys next time when they will be testing random get s ...
   Author: Vladimir Rodionov, 2012-04-04, 21:57
RE: LZ4 compression support - HBase - [mail # dev]
...LZ4 is slightly faster than snappy (from my own tests) but I really doubt t hat anyone will see performance difference in HBase (or in Hadoop)  Best regards, Vladimir Rodionov Principal...
   Author: Vladimir Rodionov, 2012-02-28, 18:20
RE: LIRS cache as an alternative to LRU cache - HBase - [mail # dev]
...afaik, existing LruBlockCache is not exactly LRU cache It utilizes more advanced algorithm to avoid cache trashing during scan ops  by dividing cache into three sub-caches (for newly ad...
   Author: Vladimir Rodionov, 2012-02-21, 17:09
RE: Scan performance on a big table as combination of multiple logic tables - HBase - [mail # dev]
...10 tables are fine. 1000 are not, especially when one does table pre-splitt ing to increase write perf.  Too many regions kill HBase.  Best regards, Vladimir Rodionov Principal Pla...
   Author: Vladimir Rodionov, 2012-02-16, 00:11
RE: Stress testing releases candidates - HBase - [mail # dev]
...+1.  Large number of regions (> 1000) per RS + heavy load - this is what we have  in our environment (staging and productions)  Its not 0.92, of course. Its 90.4 (we just u...
   Author: Vladimir Rodionov, 2012-01-06, 18:30
RE: OfflineMetaRepair? - HBase - [mail # dev]
...Unfortunately, I had to delete all data (to restore data consistency :) ) ,  therefore I can say only that, from my quick investigation, all these tables were pretty old (weeks, even &n...
   Author: Vladimir Rodionov, 2012-01-06, 05:29
RE: -ROOT- is offline - HBase - [mail # dev]
...Thanks, It was resolved after second cluster re-start. In this case - not with kill  -9 but with manual shutdown and restart of every HBase process (including Master and ZK). stop -hbas...
   Author: Vladimir Rodionov, 2012-01-05, 04:42
HBase (129)
mail # dev (66)
mail # user (60)
issue (3)
last 7 days (4)
last 30 days (5)
last 90 days (14)
last 6 months (53)
last 9 months (129)
Ted Yu (1080)
Stack (930)
Jean-Daniel Cryans (729)
Andrew Purtell (587)
stack (572)
Jonathan Hsieh (275)
Jean-Marc Spaggiari (270)
Jimmy Xiang (267)
Elliott Clark (232)
lars hofhansl (221)
Lars Hofhansl (219)
Enis Soztutar (215)
Sergey Shelukhin (195)
Doug Meil (187)
Lars George (169)
Vladimir Rodionov