[jira] [Commented] (SOLR-6730) select?replicaAffinity=(node|host) and replicaAffinity.hostPriorities support

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

[jira] [Commented] (SOLR-6730) select?replicaAffinity=(node|host) and replicaAffinity.hostPriorities support

JIRA jira@apache.org

    [ https://issues.apache.org/jira/browse/SOLR-6730?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16765888#comment-16765888 ]

Jan Høydahl commented on SOLR-6730:
-----------------------------------

Hi, commenting on this old issue, as I had a discussion with a customer yesterday regarding replica affinity to utilise caches better. They fire more than 100 requests to draw a dashboard, and you get much better cache utilisation and faster load time with replicationFactor=1 than replicationFactor=3 since you need to warm three times as many caches. Looks like the feature suggested here could help with this as well, e.g. by adding a "seed" param if you want sticky replicas for a "session". Right?

> select?replicaAffinity=(node|host) and replicaAffinity.hostPriorities support
> -----------------------------------------------------------------------------
>
>                 Key: SOLR-6730
>                 URL: https://issues.apache.org/jira/browse/SOLR-6730
>             Project: Solr
>          Issue Type: New Feature
>            Reporter: Christine Poerschke
>            Assignee: Christine Poerschke
>            Priority: Minor
>
> If no shards parameter is supplied with a select request then sub-requests will go to a random selection of live solr nodes hosting shards for the collection of interest. All sub-requests must complete before results can be collated i.e. the slowest sub-request determines how fast the search completes.
> Use of optional replicaAffinity can reduce the number of JVMs hit by a given search (the more JVMs are hit, the higher the chance of hitting a garbage collection pause in one of many JVMs). Preferentially directing requests to certain areas of the cloud can also be useful for debugging or when some replicas reside on 'faster' machines.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]