I read somewhere that we should use innodb or convert my isam to innodb in mysql. It's said that it will be good as it will be faster for database processing.
Is it right ? If yes how we convert myisam to innodb ? I think the default used in my local web server (im using WAMP) is myisam. Please tell me how to do this its because my webserver is very slow eventhough its still local.
info@elgg.org
Security issues should be reported to security@elgg.org!
©2014 the Elgg Foundation
Elgg is a registered trademark of Thematic Networks.
Cover image by RaĆ¼l Utrera is used under Creative Commons license.
Icons by Flaticon and FontAwesome.
http://www.mysqlperformanceblog.com/2009/01/12/should-you-move-from-myisam-to-innodb/
This blog post mentions converting some tables to InnoDB: http://www.arckinteractive.com/blog/2010/06/10/suggested-hardware-hosting-requirements-large-elgg-sites/
Another approach would be using an external search engine like Solr or Sphinx. The data in innodb or MyISAM tables can be exported to the search engine where a cron full index build/publish can be set up. Also, real-time index updates can be done if needed.
Well, I haven't tried this but there must be Elggers out there who already experimented with this approach, I guess :-)
It would be great if Elgg would provide integration point for external thrid party search engines.
@SkrekB:
R U out-of-date or just don't read around Elgg here enough ?;-) ;-P
Lucene and Sphinx have had PlugIns integrating into Elgg :-
I've been experimenting w/ the Sphinx one to look into upgrading to make it suitabe for 1.8.1+
@DhrupDeScoop
Zend Lucene plugin is out-of-dated (Elgg 1.6), not me :-)
As to the Sphinx Search plugin from Evan, I haven't been aware of this. So I will try it out right away .... Danke!!
So why don't you work on that Lucene PlugIn code and make it up-to-date for Egg.1.8 ? I can help guide with the technicals.
@DhrupDeScoop
No, I'd better write a Solr plugin for Elgg 1.8.1
Solr running on Tomcat has far much better performance than Zend Lucene .... So this might be one of my new year resolutions :-)
Take a look at ElasticSearch. That's what we are implementing. It really has so many benefits overs SOLR and Sphinix. Specially i like is , The way we configure ES and it's Cluster nature. Amazingly fast response. One more really cool About ElasticSearch. It is document based storage engine understand JSON.
@Cash: Thanks for references.
@iZAP Thanks for the tip. Yes, it has indeed a modern clould search architecture (e.g. node discovery, distributed, RESTful). Well, I am going to dive into more detail on this.
Btw, there is also a plugin for Amazon EC2 which allows to use AWS EC2 API for the unicast discovery mechanism as well as using S3 as a shared gateway:
https://github.com/elasticsearch/elasticsearch-cloud-aws
How far is your plugin implementation for ElasticSearch? are you intending to release this to the Elgg community?
Thanks again!
Yes, you will find so many possibilities when you dive deeper. It really do magic over EC2. We are handling that over cluster of 20 nodes.
Sorry, that is not for elgg community. We are just doing labour for our clients.
- Previous
- 1
- 2
- Next
You must log in to post replies.