Vinod Kumar Vavilapalli
Arun C Murthy
Colin Patrick McCabe
Brahma Reddy Battula
Aaron T. Myers
Tsz Wo Nicholas Sze
Uma Maheswara Rao G
Krishna Kishore Bonagiri
mail # user
mail # dev
last 7 days (2)
last 30 days (8)
last 90 days (28)
last 6 months (44)
last 9 months (219)
newest on top
oldest on top
. Results from
Loading phrases to help you
refine your search...
[HDFS-12172] Reduce EZ lookup overhead
...A number of inefficiencies exist in EZ lookups. These are amplified by frequent operations like list status. Once one encryption zone exists, all operations take the performance ...
, 2017-08-07, 03:11
[HDFS-12143] Improve performance of getting and removing inode features
...Getting a feature uses an iterator which is less performant than an indexed for loop. Feature lookups are becoming more prolific so cycles count.Removing a feature requires building a ...
, 2017-08-01, 19:14
[HDFS-12173] MiniDFSCluster cannot reliably use NameNode#stop
...Sporadic test failures occur because NameNode#stop used by the mini cluster does not properly manage the HA context's state. It directly calls HAState#exitState(context) instead of HAS...
, 2017-08-01, 16:19
[HDFS-8675] IBRs from dead DNs go into infinite loop
...If the DN sends an IBR after the NN declares it dead, the NN returns an IOE of unregistered or dead. The DN catches the IOE, ignores it, and infinitely loops spamming the NN with retri...
, 2017-07-28, 00:32
[HDFS-12137] DN dataset lock should be fair
...The dataset lock is very highly contended. The unfair nature can be especially harmful to the heartbeat handling. Under high loads, partially expose by HDFS-12136 introducing dis...
, 2017-07-25, 04:43
[HDFS-12140] Remove BPOfferService lock contention to get block pool id
...The block pool id is protected by a lock in BPOfferService. This creates excessive contention especially for xceivers threads attempting to queue IBRs and heartbeat processing. W...
, 2017-07-25, 04:42
[HDFS-12184] Avoid redundant ancestor metadata traversals for listStatus child entries.
...Creating a file status requires metadata for EC, EZ, storage policies, etc that are computed by traversing up the ancestor inodes. List status will incur the same penalties for all chi...
, 2017-07-21, 18:08
[HDFS-12174] DN commitBlockSynchronization needs delivery guarantee
...The DN's attempt to send a successful commit block synchronization to the NN is fire and forget. There are no retries. The file will linger open until the hard lease timeout. &nb...
, 2017-07-20, 21:34
[HDFS-11310] Reduce the performance impact of the balancer (trunk port)
...HDFS-7967 introduced a highly performant balancer getBlocks() query that scales to large/dense clusters. The simple design implementation depends on the triplets data structure. ...
, 2017-06-30, 00:49
[HDFS-5239] Allow FSNamesystem lock fairness to be configurable
...The fairness of the FSNamesystem#fsLock is hardcoded to true. Using an unfair lock provides a negligible increase to throughput. However this is due to bottlenecks elsewhere in t...
, 2017-06-15, 05:27
Apache Lucene, Apache Solr and all other Apache Software Foundation project and their respective logos are trademarks of the Apache Software Foundation.
Elasticsearch, Kibana, Logstash, and Beats are trademarks of Elasticsearch BV, registered in the U.S. and in other countries. This site and Sematext Group is in no way affiliated with Elasticsearch BV.
Service operated by