cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Yuki Morishita (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-4247) More-efficient serialization for IndexScanCommand and RangeSliceCommand
Date Mon, 21 May 2012 15:40:42 GMT


Yuki Morishita commented on CASSANDRA-4247:

Ah, your approach is better. :)

But I think there is a problem in serialization order. When performing range slice, first
we check whether SlicePredicate's column_names is specified, and if it is we ignore slice_range.
So we need to check column_names != null and serialize column_names first instead of range
!= null and serialize range.
> More-efficient serialization for IndexScanCommand and RangeSliceCommand
> -----------------------------------------------------------------------
>                 Key: CASSANDRA-4247
>                 URL:
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: Jonathan Ellis
>            Assignee: Yuki Morishita
>            Priority: Minor
>             Fix For: 1.2
>         Attachments: 0001-De-serialize-thrift-object-directly-from-to-stream.patch, 0002-Fix-SerializationsTest.patch
> These two classes embed a Thrift struct to simplify de/serialization. This means we serialize
to byte[] twice on the write path: once to get the size, and again to actually send the message.
> We should serialize manually like the other commands in VERSION_12.

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:!default.jspa
For more information on JIRA, see:


View raw message