Community mailing list archives
Re: Large amount of concurrent users.by
I have implemented odoo8 accounting , CRM and HRM module around 300 users. Along with that , past 6 years financial data and 1150+ employee data, and bulk customer data for CRM . Organization is having user over multi location.
IBM x3650 M4 server i have used, wth RIAD 5 technology 16GB RAM 600GB SAS HDD hexa CORE processor , keeping my database server different and application on different.
Creating a VPN and giving VPN access to every client (users ) so they get direct access to server and with proper speed.
On Tue, Jan 27, 2015 at 9:27 AM, OpenERP Master <firstname.lastname@example.org> wrote:
I could have commented better. From what I know based on experience,
All of the things Jefferey said.
Nginx reverse proxy
Separate servers (A must)Stuff he didnt mention:Put your database on an SSDUse the latest stable version of postgresql serverTune your Postgres configuration. A great guide here:
https://wiki.postgresql.org/wiki/Tuning_Your_PostgreSQL_ServerFocus on settings like:
work_memmaintenance memorymax_locks_per_transaction for larger installsshared_buffersand others in the guide.Turn off logging for when a user logs in. Each time you login it updates the database. This is a modification to the server application.If you really want to get involved, look at the slow query log. perform EXPLAIN ANALYZE on your slow queries to see the query plan.Add more indexes to fields selected a lot. Sometimes OpenERP is bad at indexing their fields for select statements
Optimize the code
Sometimes OpenERP can write inefficient code, if you know what your doing you can do some quick optimizations to really speed things up.
Things that slow down your system:
Adding calculated fields in tree view
sometimes its better to do direct SQL query, often the ORM is very inefficient with for loops in situations that are unnecessary.Hope this helpsOn Mon, Jan 26, 2015 at 6:27 PM, Jeffery <email@example.com> wrote:> >> > wrote: >> > For my o2 implementation I need to plan who built the system >> > to support 80 + concurrent connections. Does anybody have >> > experience with running a large implementation like this and >> > if so what are some steps that I should take to ensure that I >> > have adequate performance on the server. The server is a Dell >> > R 710 with 2 3. 1 ghz processor and 48 gigs of ram. I'm >> > confident that my server hardware resource is adequate however >> > from a configuration standpoint I need more information. Any >> > insight would be great. >> > >> > Thanks all >> > >> > _______________________________________________ >> > Mailing-List: https://www.odoo.com/groups/community-59have interested with this topic. i have implemented odoo instance on VM (vCPU *4 , 4G RAM) for almost 10 users. it running smoothly. there have some points need to follow up. 1. turn on workers, workers = 2 * cpu + 1 2. Faster IO for postgresql 3. more memory for postgresql, also need to tune memory usage for db 4. change the log_level from info to warning or critical, to reduce the logger writing disk 5. use nginx for proxy 6. separate the db and application servers from the same server, as postgresql need more IO and odoo application need more CPU 7. load balance for application server, and or postgresql cluster. hope those tips should help you. On Mon, Jan 26, 2015 at 11:22 PM, OpenERP Master <firstname.lastname@example.org> wrote: > Hi, > > I'm going to start on a project with 200+ users and over 1.2 million > products. Will let you know what we did. > > On Mon, Jan 26, 2015 at 8:42 AM, Dave Ellison <email@example.com> wrote: >> >> Hi, >> >> Like the presentation, will store this link for the future >> >> On Mon, 2015-01-26 at 12:57 +0000, Gustavo Orrillo wrote: >> > this presentation will help you >> > >> > >> > http://es.slideshare.net/openobject/performance2014-35689113 >> > >> > >> > On Mon, Jan 26, 2015 at 10:13 AM, John Pia Jr <firstname.lastname@example.org-- Jeffery -kunshan>> > Post to: mailto:email@example.com >> > Unsubscribe: https://www.odoo.com/groups?unsubscribe >> > >> > >> > >> > >> > >> > _______________________________________________ >> > Mailing-List: https://www.odoo.com/groups/community-59 >> > Post to: mailto:firstname.lastname@example.org >> > Unsubscribe: https://www.odoo.com/groups?unsubscribe >> > >> > >> > >> >> _______________________________________________ >> Mailing-List: https://www.odoo.com/groups/community-59 >> Post to: mailto:email@example.com >> Unsubscribe: https://www.odoo.com/groups?unsubscribe > > > _______________________________________________ > Mailing-List: https://www.odoo.com/groups/community-59 > Post to: mailto:firstname.lastname@example.org > Unsubscribe: https://www.odoo.com/groups?unsubscribe
*Please Don't print this email unless you really need to "Save Earth and Save Environment"