[
https://issues.apache.org/jira/browse/IGNITE-4908?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16072740#comment-16072740
]
Andrew Mashenkov commented on IGNITE-4908:
------------------------------------------
[~sharpler],
Yes, Ignite.reentrantLock() is ignite datastructure that stored in system
cache and uses a pessimistic repeatable read transaction for updates.
IgniteCache.lock() looks like avoids a heavy transaction, but I'm not sure it
is just a local lock. I would think it is distributed lock.
How many nodes grid do you try?
What guarantees do you mean? Fairness? Unfair locks seems to have same
performance as it also uses transactions.
For now, user doesn't have a distributed lock with relaxed guaranties that
would work as IgniteCache.lock(), but do not affect cache operations.
So, I create the ticket investigate a reason of slowdown and check if there is
a way to speed up datastructure locks.
Is there any reason why we can't have an additional implementation for
datastructure lock based on IgniteCache.lock() with same speed an guarantee?
> Ignite.reentrantLock looks much slower than IgniteCache.lock.
> -------------------------------------------------------------
>
> Key: IGNITE-4908
> URL: https://issues.apache.org/jira/browse/IGNITE-4908
> Project: Ignite
> Issue Type: Improvement
> Components: data structures
> Affects Versions: 1.8
> Reporter: Andrew Mashenkov
> Assignee: Alexander Menshikov
> Fix For: 2.1
>
>
> We should make a benchmark and investigate this issue.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
