[
https://issues.apache.org/jira/browse/IGNITE-26885?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Pavel Tupitsyn updated IGNITE-26885:
------------------------------------
Description:
As a user, I'd like to know if partition awareness works efficiently (key-based
requests go directly to the primary replica) on the server side - whether
requests are rerouted often.
See also IGNITE-26884 (same metric on the client side).
was:
As a user, I'd like to know if partition awareness works efficiently (key-based
requests go directly to the primary replica).
In *ReliableChannel* we have *preferredNodeName*. If a connection to that node
is available, it is a "hit", otherwise a "miss".
Note that partition assignment can be out of date on the client at certain
times, so we don't actually know if a given request was rerouted on the server,
but the client-side metric is still useful.
> Add server-side partition awareness hit/miss metrics
> ----------------------------------------------------
>
> Key: IGNITE-26885
> URL: https://issues.apache.org/jira/browse/IGNITE-26885
> Project: Ignite
> Issue Type: Improvement
> Components: thin clients ai3
> Reporter: Pavel Tupitsyn
> Priority: Major
> Labels: ignite-3
> Fix For: 3.2
>
>
> As a user, I'd like to know if partition awareness works efficiently
> (key-based requests go directly to the primary replica) on the server side -
> whether requests are rerouted often.
> See also IGNITE-26884 (same metric on the client side).
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
