[TYPO3-Solr] seperate indexing and search

[Johann Höchtl]

Hi Robert,

this article with deadly sins is from 2010. Since then the Solr index structure has evolved dramatically and it’s not longer that big an issue to have indexing and searching on the same node. 
If you are using TYPO3 with Solr usually the amount of data you have inside your solr-index is way to less to make this separation necessary.
You need to think about this if you have millions of records and refresh them every hour and have more than 100 Req/s on your solr server.

At the moment this is not possible out of the box, but you could simply modify the the Apache_Solr_Service class which constructs the updateUrl to use a different host there for the updateUrl (of course add setting params etc.).

But if you need more performance than your solr server has, install Solr-Cloud which takes care of all this and has a single endpoint.
SolrCloud scales nearly linear and you index and search on the entire distributed system. You also gain availability besides performance ^^

Best regards
Hans 

> Am 24.10.2014 um 08:38 schrieb Robert W <rewebit at gmail.com>:
> 
> Hello,
> 
> In reading online, there was a reccomendation for seperating search and indexing to two different nodes (ex. https://lucidworks.com/blog/the-seven-deadly-sins-of-solr/). I have not seen in the documentation or in searching the TYPO3 forum, if this is possible. Can the Solr extension be configured for seperate search and indexing machines? 
> Currently we are putting together our production environment and are trying to see how we want to set this up. Thank you for your help.
> _______________________________________________
> TYPO3-project-solr mailing list
> TYPO3-project-solr at lists.typo3.org
> http://lists.typo3.org/cgi-bin/mailman/listinfo/typo3-project-solr