Clustering

Previous topic

Developer

Next topic

Author	Message
Björn Dieding@xrow.de	Tuesday 12 August 2003 5:48:17 pm Wondering I was wondering how you archive proper clustering. AS we know Mysql can be setup as a clustered database. Then you could propably use round robin for the to address each application server. Cache and variations can be biuld on each application server. But how about the original files you uploaded to the system shared through out the system? Would you recommend to implement NSF there and share it across the cluster? Or would it be wiser to try to safe the original binary files into the database too? I have seen MYSQL databases with up to 1 GB binary data and that database was also performing well. Does anybody has any idea? Looking for a new job? http://www.xrow.com/xrow-GmbH/Jobs Looking for hosting? http://hostingezpublish.com ----------------------------------------------------------------------------- GMT +01:00 Hannover, Germany Web: http://www.xrow.com/
Tony Wood	Wednesday 13 August 2003 1:33:21 am Hi, Are you after true clustering if so what clustering platform are you planning on using? If not try using rsync for user files, but not system files, even a 1gb NFS direct crossover will give speed that is too slow for main usage? How many users are you looking to support? in total and concurrently? tony Tony Wood : twitter.com/tonywood Vision with Technology Experts in eZ Publish consulting & development Power to the Editor! Free eZ Training : http://www.VisionWT.com/training eZ Future Podcast : http://www.VisionWT.com/eZ-Future
Björn Dieding@xrow.de	Monday 18 August 2003 2:44:05 pm Thanks tony for your response. We currently have up to 1,5 million hits a day on non static pages and we are currently using just a good but single server. I have no idea about proper clustering systems yet (OS based). That why I would like to know how others would implement it. Did I get you right, in recommending a NFS share for storing the origional variations? Looking for a new job? http://www.xrow.com/xrow-GmbH/Jobs Looking for hosting? http://hostingezpublish.com ----------------------------------------------------------------------------- GMT +01:00 Hannover, Germany Web: http://www.xrow.com/

Author

Message

Björn Dieding@xrow.de

Tuesday 12 August 2003 5:48:17 pm

Wondering

I was wondering how you archive proper clustering. AS we know Mysql can be setup as a clustered database. Then you could propably use round robin for the to address each application server. Cache and variations can be biuld on each application server. But how about the original files you uploaded to the system shared through out the system? Would you recommend to implement NSF there and share it across the cluster? Or would it be wiser to try to safe the original binary files into the database too? I have seen MYSQL databases with up to 1 GB binary data and that database was also performing well.

Does anybody has any idea?

Looking for a new job? http://www.xrow.com/xrow-GmbH/Jobs
Looking for hosting? http://hostingezpublish.com
-----------------------------------------------------------------------------
GMT +01:00 Hannover, Germany
Web: http://www.xrow.com/

Tony Wood

Wednesday 13 August 2003 1:33:21 am

Hi,

Are you after true clustering if so what clustering platform are you planning on using?
If not try using rsync for user files, but not system files, even a 1gb NFS direct crossover will give speed that is too slow for main usage?

How many users are you looking to support? in total and concurrently?

tony

Tony Wood : twitter.com/tonywood
Vision with Technology
Experts in eZ Publish consulting & development

Power to the Editor!

Free eZ Training : http://www.VisionWT.com/training
eZ Future Podcast : http://www.VisionWT.com/eZ-Future

Björn Dieding@xrow.de

Monday 18 August 2003 2:44:05 pm

Thanks tony for your response.

We currently have up to 1,5 million hits a day on non static pages and we are currently using just a good but single server. I have no idea about proper clustering systems yet (OS based). That why I would like to know how others would implement it.

Did I get you right, in recommending a NFS share for storing the origional variations?

eZ debug

Timing:	Jan 18 2025 19:29:01
Script start
Timing:	Jan 18 2025 19:29:01
Module start 'layout'
Timing:	Jan 18 2025 19:29:01
Module start 'content'
Timing:	Jan 18 2025 19:29:01
Module end 'content'
Timing:	Jan 18 2025 19:29:01
Script end

Main resources:

Total runtime	0.0143 sec
Peak memory usage	2,048.0000 KB
Database Queries	3

Timing points:

Checkpoint	Start (sec)	Duration (sec)	Memory at start (KB)	Memory used (KB)
Script start	0.0000	0.0053	587.7891	152.6094
Module start 'layout'	0.0053	0.0029	740.3984	39.3984
Module start 'content'	0.0083	0.0043	779.7969	93.2891
Module end 'content'	0.0125	0.0017	873.0859	34.3047
Script end	0.0142		907.3906

Time accumulators:

Accumulator	Duration (sec)	Duration (%)	Count	Average (sec)
Ini load
Load cache	0.0025	17.8372	14	0.0002
Check MTime	0.0011	7.6813	14	0.0001
Mysql Total
Database connection	0.0007	5.1130	1	0.0007
Mysqli_queries	0.0023	16.4245	3	0.0008
Looping result	0.0000	0.1052	1	0.0000
Template Total	0.0015	10.2	1	0.0015
Template load	0.0007	5.2249	1	0.0007
Template processing	0.0007	4.9578	1	0.0007
Override
Cache load	0.0005	3.3547	1	0.0005
General
dbfile	0.0003	1.8903	8	0.0000
String conversion	0.0000	0.0584	4	0.0000
Note: percentages do not add up to 100% because some accumulators overlap

Templates used to render the page:

Usage	Requested template	Template	Template loaded	Edit	Override
1	print_pagelayout.tpl	<No override>	extension/community/design/community/templates/print_pagelayout.tpl
Number of times templates used: 1 Number of unique templates used: 1

Time used to render debug report: 0.0001 secs