[
https://issues.apache.org/jira/browse/CASSANDRA-9499?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14588647#comment-14588647
]
Ariel Weisberg commented on CASSANDRA-9499:
-------------------------------------------
I got the readVInt optimization where there is no loop working.
What are we looking to accomplish with a different encoding?
* Bias towards positive numbers
* More efficient length extension (currently the first byte is dropped) for > 1
byte values
* Simpler/faster implementation
I just bent over backwards to remove branching, loops, instructions etc. If we
change encodings are we going to lose that ground? Is what we would get from
the first two attributes more important?
Do we want to do just one, say more efficient length extension?
> Introduce writeVInt method to DataOutputStreamPlus
> --------------------------------------------------
>
> Key: CASSANDRA-9499
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9499
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Benedict
> Assignee: Ariel Weisberg
> Priority: Minor
> Fix For: 3.0 beta 1
>
>
> CASSANDRA-8099 really could do with a writeVInt method, for both fixing
> CASSANDRA-9498 but also efficiently encoding timestamp/deletion deltas. It
> should be possible to make an especially efficient implementation against
> BufferedDataOutputStreamPlus.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
