Sizing p7 systems


 
Thread Tools Search this Thread
Operating Systems AIX Sizing p7 systems
# 1  
Old 11-03-2010
Sizing p7 systems

Hi,

I am currently given the dubious task Smilie to make an assumption on sizing our current p5 environment in relation to new p7 machines. Is there any kind of a vague factor how to calculate this?
I found some official IBM benchmark results with long lists of different Power machines and their scores but since I do not know the benchmarks themselves they were kinda useless for me.
Is there a simple kind of formula how to guess what number of p7 CPUs you might need to compensate a number of p5 CPUs?

Thanks in advance.

Last edited by zaxxon; 11-04-2010 at 06:10 AM.. Reason: typo in subject
# 2  
Old 11-03-2010
You might start with specInt ratings, unless your work is highly floating point calculation intensive.
# 3  
Old 11-03-2010
Hi Zaxxon,

what kind of apps are you running (DBs, Websphere ...) and are your p5 virtualized or capped Smilie And will your cpu pool be generous or rather low.

I am doing the a lot of sizing for my company - at the moment we are consolidating 38 small p5 frames (550/570ies) into 4 780ies ... what I find is that from a pure performance perspective a few rules of thumb still work perfectly fine - we have generous pools so I do not really care about cpu entitlements when I do the initial setup. If I have a sybase DB, I use one virtual per engine (backup server and repserver count as engines) + 1 as count for virtual cpus, on Oracle I am checking how many parallel processes (not connections) I have running on the server I want to migrate - that will be the amount of parallel threads I want to allow on the new box too - 1 virtual = 4 threads + 1 virtual for the OS. My virtuals are generally worth 0.1 physical cpu - as they can grow to a physical cpu during peaks that is just fine - what is not needed goes back into the pool. Monitor about 4 weeks after go live closely and than amend the amount of entitled cpu to what you usually need (so in average, NOT during peaks) and that is what they get permanently. Regarding memory - well its pretty much the same.
For websphere boxes we usually go with half as many virtuals as we were using on p5. Again - monitor closely and amend if required.
In my experience you have usually a lot of cpus in the pool and most frames do not have the peak times of all their lpars same time so you should usually be good.

For DBs you can go if you like as well another way: p5 to p6 = 3:2, p5 to p7 = 7:4 - that is what our engineering came up with - they assume high frame usage and small pool Smilie

I have the official IBM numbers in the office - can post tomorrow ...

Hope that helps
regards
zxmaus
This User Gave Thanks to zxmaus For This Post:
# 4  
Old 11-04-2010
Thanks for the infos so far.

@DGPicket
For those specInt ratings I have no clue as it sounds a bit abstract to me; will have to read about it to be able to interpret them.

@zxmaus
We have a lot different partitions, some capped, some uncapped and apart from saving licenses there is often no real reason why they are capped or not, some shared some dedicated. I will keep it in mind but I am maybe not the one who is deciding this and people here are often resistant to advices.
Application wise we are having a lot of Oracle DBs, a few Informix DBs and some boxes with Weblogic, which should be comparable to Websphere; there are also some minor applications but they are not that hungry.
For the generousity of pools, I do not expect much here. But that is another topic Smilie

I would really appreciate those official IBM infos, thanks a lot in forward ^^
# 5  
Old 11-04-2010
Hi Zaxxon,

as promised ... the official numbers from IBM here - let me know if you need help to understand this Smilie
# 6  
Old 11-04-2010
Neat! Remember that for the same total nominal spec-int or whatever, fewer processors is really faster. You only get full speed from the first, and then for each additional one sharing the same RAM, bus's, L2cache, L1cache, you get less. 4 cores of 20 > 8 cores of 10.
# 7  
Old 11-05-2010
The standard for doing this work is to compare the rPerf number in the link that zxmaus gave you. This is an IBM standard where a p640 has a value of 1. This is an analytical model of commercial workloads rather than fancy benchmarks so has good application in the field.

Note that it doesn't look at I/O (Network and disk). Also your IBM representative (or BP) should be able to provide you with assistance that your strategy for consolidation is correct.

Don't forget memory overheads for LPARS, VIO etc etc.
Login or Register to Ask a Question

Previous Thread | Next Thread

7 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

Using symbolic link for database MySQL in CentOS, not update sizing

I have no idea what I should set the topic here ==' This is my story, please you there kindly help me I'm quite newbie for this. ================================== My host server is CentOS, I spared 9.9GB for /var path that used by MySQL and...It's full because of heavy load traffic, then... (1 Reply)
Discussion started by: Kapom
1 Replies

2. What is on Your Mind?

From Systems Admin to Systems Eng.

I have been wondering how do Systems Administrators do the jump into Systems Engineering? Is it only a matter of time and experience or could I actually help myself get there? Opinions? Books I could read? Thanks a lot for your help! (0 Replies)
Discussion started by: svalenciatech
0 Replies

3. Solaris

Sun hardware sizing

Hi, Never worked with Sun, but I have been presented to make a decision about Sun server hardware, since the application which we'll be running is not so popular and you guys might not have idea, for reference I can tell you our competitor is running same application (business volume 10 times... (1 Reply)
Discussion started by: tayyabq8
1 Replies

4. Linux

System sizing for X sessions

Hello all. I've been tasked with building a system that will have up to 50 concurrent users connected and using an X session running firefox. Is there some kind of standard sizing model available that will tell me what kind of network/CPU utilization I will be looking at? Or even better some... (1 Reply)
Discussion started by: ZekesGarage
1 Replies

5. Shell Programming and Scripting

Kinda a noob: Automatic window sizing and placement

I am attempting to create a script that runs automatically upon logging in and opens and places windows in appropriate places. I have the script running such that it starts during login, but I cannot get things how and where I want them. This should be relatively simple, I just can't figure it out... (7 Replies)
Discussion started by: wydileie
7 Replies

6. AIX

Sizing Mouse Pointer

On AIX machines 4.3, 5.1, 5.2 and 5.3 am having trouble seeing the mouse pointer in CDE. Found this on web To change the cursor to a large red X, run the following command . #xsetroot -cursor_name X-cursor -bg red Could someone kindly give me the command line wording to undo the above... (2 Replies)
Discussion started by: farl
2 Replies

7. Gentoo

Properly Sizing an x86 Server for Internet DNS?

Where I work, we have to very old Alpha boxes running OpenVMS 7. They also have Multinet and are using the BIND component for DNS services. We are planning on retiring those boxes and replacing them with x86 servers running Linux. I've decided to go with Gentoo Linux for this and I've inherited... (3 Replies)
Discussion started by: deckard
3 Replies
Login or Register to Ask a Question