[
https://issues.apache.org/jira/browse/IGNITE-1665?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14967335#comment-14967335
]
Pavel Tupitsyn edited comment on IGNITE-1665 at 10/21/15 4:59 PM:
-------------------------------------------------------------------
* type id check restored, writer logic is now intact
* reader and writer type structure are now separate:
** Read and write order may be different
** Reader always uses typeId=0 so it won't clash with writer checks
Reader type validation ticket: https://issues.apache.org/jira/browse/IGNITE-1760
was (Author: ptupitsyn):
* type id check restored, writer logic is now intact
* reader and writer type structure are now separate:
** Read and write order may be different
** Reader ignores type id checks and won't clash with writer checks
Reader type validation ticket: https://issues.apache.org/jira/browse/IGNITE-1760
> .Net: Cache field IDs when deserializing objects.
> -------------------------------------------------
>
> Key: IGNITE-1665
> URL: https://issues.apache.org/jira/browse/IGNITE-1665
> Project: Ignite
> Issue Type: Task
> Components: interop
> Affects Versions: ignite-1.4
> Reporter: Vladimir Ozerov
> Assignee: Pavel Tupitsyn
> Priority: Critical
> Fix For: 1.5
>
>
> Profiliing deserialization of test "Address" object shows that we spend
> considerable amount of time (10-20%) inside field ID calculation.
> We already optimized this for writes using serialization "paths" concept
> where field hashing is replaced with strings reference equality checks. The
> same thing should be done for reads.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
