[
https://issues.apache.org/jira/browse/IGNITE-29169?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18132399#comment-18132399
]
Ignite TC Bot commented on IGNITE-29169:
----------------------------------------
{panel:title=Branch: [pull/13706/head] Base: [master] : No blockers
found!|borderStyle=dashed|borderColor=#ccc|titleBGColor=#D6F7C1}{panel}
{panel:title=Branch: [pull/13706/head] Base: [master] : No new tests
found!|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}{panel}
[TeamCity *--> Run :: All*
Results|https://ci2.ignite.apache.org/viewLog.html?buildId=9442968&buildTypeId=IgniteTests24Java8_RunAll]
{color:#ffffff}tcbot-analysis-comment chainBuildId=9442968
rerunBuildIds=9442970,9442972{color}
> Use CopyOnWriteIntMap for rarely updated maps on cache hot paths
> ----------------------------------------------------------------
>
> Key: IGNITE-29169
> URL: https://issues.apache.org/jira/browse/IGNITE-29169
> Project: Ignite
> Issue Type: Improvement
> Reporter: Nikita Amelchev
> Assignee: Nikita Amelchev
> Priority: Minor
> Fix For: 2.19
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Several int-keyed maps that are read on the cache operation path but updated
> only when a cache or a cache group starts or stops still pay for concurrent
> writes on every read: the per-partition entry and size maps of shared cache
> groups take a read-write lock on each access, {{DataRegionMetricsImpl}} keeps
> a hand-written copy-on-write map guarded by its own lock, and the affinity
> key field cache boxes the type ID on every lookup. This is exactly the
> read-mostly case {{CopyOnWriteIntMap}} from IGNITE-29099 was introduced for,
> so these maps should use it instead of their own locking and copying.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
