[
https://issues.apache.org/jira/browse/IGNITE-28965?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Vladimir Steshin updated IGNITE-28965:
--------------------------------------
Description: `{_}TcpDiscoveryImpl{_}` caches its `{_}ClusterMetrics{_}` and
`{_}CacheMetrics{_}`. Or we set and store snapshot of `{_}ClusterMetrics{_}`
only to join cluster. Despite this, `{_}TcpDiscoveryImpl{_}` has
`{_}DiscoveryMetricsProvider{_}` at least for local node instances. That looks
like we have some extra memory consumption (for local node?):
`{_}DiscoveryMetricsProvider{_}` provides actual metrics on demand and we keep
the metrics snapshots simultaneously. (was: `{_}TcpDiscoveryImpl{_}` caches
its `{_}ClusterMetrics{_}` and `{_}CacheMetrics{_}`. Or we set and store
snapshot of `{_}ClusterMetrics{_}` only to join cluster. Despite this,
`{_}TcpDiscoveryImpl{_}` has `{_}DiscoveryMetricsProvider{_}` at least for
local node instances. That looks like we have some extra memory consumption:
`{_}DiscoveryMetricsProvider{_}` provides actual metrics on demand and we keep
the metrics snapshots simultaneously.)
> Investigate cached metrics in TcpDiscoveryNode
> ----------------------------------------------
>
> Key: IGNITE-28965
> URL: https://issues.apache.org/jira/browse/IGNITE-28965
> Project: Ignite
> Issue Type: Task
> Reporter: Vladimir Steshin
> Priority: Minor
>
> `{_}TcpDiscoveryImpl{_}` caches its `{_}ClusterMetrics{_}` and
> `{_}CacheMetrics{_}`. Or we set and store snapshot of `{_}ClusterMetrics{_}`
> only to join cluster. Despite this, `{_}TcpDiscoveryImpl{_}` has
> `{_}DiscoveryMetricsProvider{_}` at least for local node instances. That
> looks like we have some extra memory consumption (for local node?):
> `{_}DiscoveryMetricsProvider{_}` provides actual metrics on demand and we
> keep the metrics snapshots simultaneously.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
