Frank,

I think there have been improvements in virtualization as well. R&D and 
services use this heavily. If you have well maintained fast servers virtual 
or not they should be fine. If you have a farm on old hardware or where 
there are more images than underlying resources swapping could create lag. 
Do they have reporting tools they'll run on performance? It would be nice 
if you had RDP to Windows & IIS access otherwise if properly scaled you 
should be better off.

This is the 11.0 Whitepaper on performance from 18 months ago. 11.2 is 
about to be released. I'd shoot for 11.1 HF4 or later for any upgrades.
https://knowledge.opentext.com/knowledge/cs.dll?func=ll&objaction=overview&objid=27439570

I got a chance to talk with Uli from KIT (who runs 650 projects), and was 
the primary subject of the above whitepaper, at Enterprise World this year 
and he shared that he recently ran 86 users on one server (24 core I 
believe) instead of his usual 4 boxes for editing and publication. Doing 
that he logged in and saw no noticeable lag. If you reach out to the RD 
User Group I'm sure they can put you in touch with him to compare notes.

You can also do MS Server 2012 and MS SQL 2012 in 11.1+

The other reason to go v11.1+ is the browsers. WSM is setting the stage to 
turn off support for older browsers soon because new browsers have better 
JavaScript performance which is key to the editors experience.

Best,
Tim

On Thursday, December 5, 2013 10:22:31 AM UTC-5, Frank Leja wrote:
>
> Hello,
>
> we have a V9.0.1.49 CMS cluster with almost complete physical machines 
> (except the publishing server) and want to jump directly to V11.1. 
>
> Currently our cluster consists of these hardware/software.
>   Name kfdcrd10 kfdcrd11 kfdcrd20 kswfrd03  Role author author publish 
> database  Type physical physical VM physical  CPU [email protected]<javascript:> 
> [email protected] <javascript:> [email protected] <javascript:> 
> [email protected]<javascript:>  
> Cores 8 8 8 4  RAM 8GB 8GB 16GB 8GB  OS Windows Server 2003 SP2 Windows 
> Server 2003 SP2 Windows Server 2003 SP2 Windows Server 2003 SP2  Host n/a 
> n/a IBM X3850 M2 n/a  Host 2014 n/a n/a IBM X3650 M4 n/a 
> My operations department force me to use only virtual servers, even the 
> database. We tried some years ago a virtualisation, but it failed because 
> of the slow file I/O on the VMs. 
>
> I'm sure, that the V11 "best RedDot CMS ever" will be much faster if we 
> keep our new cluster on physical hardware (slightly modernized). I want to 
> bring this better performance to the desk of my authors, which are working 
> worldwide. They have to learn the new RD UI - better performance could be a 
> driver to accept this effort.
>
> Because of my experience with the failed virtualization I would like to 
> have your experience or recommendation, if a V9 => V11 performance 
> improvement will be eaten by the virtualization? 
>
> 1) Is it wise to virtualize the DB server as well?
> 2) Could the both authoring servers with V11 virtualized, because they are 
> now in .NET, what works for publishing yet in V9 satisfying?
>
> Some of my pros and cons:
>
>    - pro: our database is a very old MS SQL Server 2000 - the new 2008 
>    version is better optimized for virtualization I heard
>    - pro: V11 uses .NET with better multi-user support
>    - pro: file I/O for logging seems to be better configurable in V11 
>    (not all or nothing anymore as in V9)
>    - pro: we will get a new X3650 M4 host 
>    - pro: VMWare itself is become better
>    - con: all (VM, IIS, DB, OS) is outsourced and I have no view into and 
>    (of course no) control on the ESX host at all
>    - con: I'm the RD admin only our operating department is separated, 
>    but they have limited knowledge about RD
>    - con: heard from other VM applications, that the performance changes 
>    dramatically: same job can take 5s or 4min
>    - con: we make very heavy usage of RQL commands via 
> jRQL<http://jrql.wordpress.com/>; 
>    one of our authors creates much more stress to the servers as normal (and 
>    we have lots of batch jobs running on author machines)
>    - con: unsure how to make a realistic performance measurement and 
>    comparision of the old V9 and new V11.1 environment
>
> Would appreciate your assessment.
>
> Frank Leja
>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"RedDot CMS Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/reddot-cms-users.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to