[ 
https://issues.apache.org/jira/browse/CASSANDRA-14556?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16543733#comment-16543733
 ] 

ASF GitHub Bot commented on CASSANDRA-14556:
--------------------------------------------

Github user aweisberg commented on a diff in the pull request:

    https://github.com/apache/cassandra/pull/239#discussion_r202476142
  
    --- Diff: src/java/org/apache/cassandra/db/DiskBoundaries.java ---
    @@ -129,4 +129,19 @@ public int 
getBoundariesFromSSTableDirectory(Descriptor descriptor)
         {
             return directories.get(getDiskIndex(sstable));
         }
    +
    +    public Directories.DataDirectory getCorrectDiskForKey(DecoratedKey key)
    +    {
    +        if (positions == null)
    +            return null;
    +
    +        return directories.get(getDiskIndex(key));
    +    }
    +
    +    private int getDiskIndex(DecoratedKey key)
    +    {
    +        int pos = Collections.binarySearch(positions, key);
    --- End diff --
    
    OK so this is insertion point, and then -1 because the of 0 based indexing? 
Can you unit test this just to make sure that it find the correct spot?
    
    I'm just not following the -1 I guess because Collections.binarySearch says 
insertion point -1 is what is returned. What if the key happens to be out of 
range of the ranges this node owns? This would go out of bounds?


> Optimize streaming path in Cassandra
> ------------------------------------
>
>                 Key: CASSANDRA-14556
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-14556
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Streaming and Messaging
>            Reporter: Dinesh Joshi
>            Assignee: Dinesh Joshi
>            Priority: Major
>              Labels: Performance
>             Fix For: 4.x
>
>
> During streaming, Cassandra reifies the sstables into objects. This creates 
> unnecessary garbage and slows down the whole streaming process as some 
> sstables can be transferred as a whole file rather than individual 
> partitions. The objective of the ticket is to detect when a whole sstable can 
> be transferred and skip the object reification. We can also use a zero-copy 
> path to avoid bringing data into user-space on both sending and receiving 
> side.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to