hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Zheng Shao (JIRA)" <>
Subject [jira] Created: (HIVE-535) Memory-efficient hash-based Aggregation
Date Tue, 02 Jun 2009 20:03:07 GMT
Memory-efficient hash-based Aggregation

                 Key: HIVE-535
             Project: Hadoop Hive
          Issue Type: Improvement
    Affects Versions: 0.4.0
            Reporter: Zheng Shao

Currently there are a lot of memory overhead in the hash-based aggregation in GroupByOperator.
The net result is that GroupByOperator won't be able to store many entries in its HashTable,
and flushes frequently, and won't be able to achieve very good partial aggregation result.

Here are some initial thoughts (some of them are from Joydeep long time ago):

A1. Serialize the key of the HashTable. This will eliminate the 16-byte per-object overhead
of Java in keys (depending on how many objects there are in the key, the saving can be substantial).
A2. Use more memory-efficient hash tables - java.util.HashMap has about 64 bytes of overhead
per entry.
A3. Use primitive array to store aggregation results. Basically, the UDAF should manage the
array of aggregation results, so UDAFCount should manage a long[], UDAFAvg should manage a
double[] and a long[]. The external code should pass an index to iterate/merge/terminal an
aggregation result. This will eliminate the 16-byte per-object overhead of Java.

More ideas are welcome.

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

View raw message