hadoop-pig-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Pi Song (JIRA)" <j...@apache.org>
Subject [jira] Commented: (PIG-170) Memory manager spills bags in the wrong order
Date Fri, 28 Mar 2008 13:44:24 GMT

    [ https://issues.apache.org/jira/browse/PIG-170?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12583025#action_12583025
] 

Pi Song commented on PIG-170:
-----------------------------

In certain scenarios, for example, a lot of small intermediate bags are used to calculate
something and outputs are stored in a big bag. The big bag will become the first node after
{noformat}
Collections.sort(spillables, new Comparator<WeakReference<Spillable>>()
{noformat}

 and the below code will no longer work.
{noformat}
    public void registerSpillable(Spillable s) {
        synchronized(spillables) {
            // Cleaing the entire list is too expensive.  Just trim off the front while
            // we can.
            WeakReference<Spillable> first = spillables.peek();
            while (first != null && first.get() == null) {
                spillables.remove();
                first = spillables.peek();
            }
            spillables.add(new WeakReference<Spillable>(s));
        }
    }
{noformat}
I

> Memory manager spills bags in the wrong order
> ---------------------------------------------
>
>                 Key: PIG-170
>                 URL: https://issues.apache.org/jira/browse/PIG-170
>             Project: Pig
>          Issue Type: Bug
>            Reporter: Olga Natkovich
>            Assignee: Amir Youssefi
>         Attachments: PIG-170_0_20080327.patch
>
>
> For optimal performance, we want to spill the largest bags first. This is not what is
happening right now and could be causing some of our memory issues.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message