[ 
https://issues.apache.org/jira/browse/HBASE-5010?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13174574#comment-13174574
 ] 

Phabricator commented on HBASE-5010:
------------------------------------

Kannan has commented on the revision "[jira] [HBASE-5010] [89-fb] Filter HFiles 
based on TTL".

  Mikhail: Nice work, and unit test. Compaction related comment inline.

INLINE COMMENTS
  src/main/java/org/apache/hadoop/hbase/regionserver/StoreScanner.java:133 
pre-existing issue: this constructor is no longer just for major compactions. 
Minor compactions also use this.
  src/main/java/org/apache/hadoop/hbase/regionserver/StoreScanner.java:154 If 
we do a similar check in this constructor, we would get the same optimization 
for compactions also.

  As we talked about offline, if the entire file has expired data, then we can 
avoid adding the scanner to  the KeyValueHeap below. So for CFs which have 
routinely expiring data due to TTL, compactions would have to read a lot less 
data too or could essentially turn into feather-weight ops which just delete 
unnecessary/old files.

REVISION DETAIL
  https://reviews.facebook.net/D909

                
> Filter HFiles based on TTL
> --------------------------
>
>                 Key: HBASE-5010
>                 URL: https://issues.apache.org/jira/browse/HBASE-5010
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Mikhail Bautin
>            Assignee: Mikhail Bautin
>         Attachments: D909.1.patch
>
>
> In ScanWildcardColumnTracker we have
> {code:java}
>  
>   this.oldestStamp = EnvironmentEdgeManager.currentTimeMillis() - ttl;
>   ...
>   private boolean isExpired(long timestamp) {
>     return timestamp < oldestStamp;
>   }
> {code}
> but this time range filtering does not participate in HFile selection. In one 
> real case this caused next() calls to time out because all KVs in a table got 
> expired, but next() had to iterate over the whole table to find that out. We 
> should be able to filter out those HFiles right away. I think a reasonable 
> approach is to add a "default timerange filter" to every scan for a CF with a 
> finite TTL and utilize existing filtering in 
> StoreFile.Reader.passesTimerangeFilter.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Reply via email to