Personal computing discussed
Moderators: renee, morphine, Steel
TR Forums Iometer Workload Patterns
-----------------------------------
(First created: 2009/11/24)
Files:
1. trforums_single.icf - all patterns with 1 worker (single CPU core)
2. trforums_multi.icf - all patterns with multiple workers (1 per CPU core)
Access specifications:
1. Idle (original)
2. Default (original)
3. All in one (original)
4. File Server - Intel/StorageReview.com
http://www.storagereview.com/articles/200003/20000313OSandBM_5.html
5. Web Server - StorageReview.com
http://www.storagereview.com/articles/200111/20011109Renaissance_4.html?page=0,7
6. Database - Read - Intel/StorageReview.com
http://www.storagereview.com/articles/200003/20000313OSandBM_5.html
7. Database - Write - Mirror image of "Database - Read"
8. Workstation - SR - StorageReview.com
http://ixbtlabs.com/articles/hddide2k1feb/iometer.html
9. Workstation - X-bit labs (GReY)
http://www.xbitlabs.com/articles/storage/display/sil-3124_3.html#sect0
10. Streaming Writes - THG
http://www.tomshardware.com/reviews/x25-e-ssd-performance,2365-7.html
11. Streaming Reads - THG
http://www.tomshardware.com/reviews/x25-e-ssd-performance,2365-7.html
Test setups:
1. Run time: 10 minutes
2. Ramp up time: 20 seconds
3. Outstanding I/O's: 1, 2, 4, 8, 16, 32, 64, 128, 256 (9 tests)
4. Number of workers: controlled by file
Version 2006.07.27
'TEST SETUP ====================================================================
'Test Description
TR Forums - Single Core CPU
'Run Time
' hours minutes seconds
0 10 0
'Ramp Up Time (s)
20
'Default Disk Workers to Spawn
1
'Default Network Workers to Spawn
NUMBER_OF_CPUS
'Record Results
ALL
'Worker Cycling
' start step step type
1 1 LINEAR
'Disk Cycling
' start step step type
1 1 LINEAR
'Queue Depth Cycling
' start end step step type
1 256 2 EXPONENTIAL
'Test Type
CYCLE_OUTSTANDING_IOS
'END test setup
'ACCESS SPECIFICATIONS =========================================================
'Access specification name,default assignment
Default,NONE
'size,% of size,% reads,% random,delay,burst,align,reply
2048,100,67,100,0,1,0,0
'Access specification name,default assignment
All in one,NONE
'size,% of size,% reads,% random,delay,burst,align,reply
512,5,100,0,0,1,0,0
512,5,75,0,0,1,0,0
512,5,50,0,0,1,0,0
512,5,25,0,0,1,0,0
512,5,0,0,0,1,0,0
4096,5,100,0,0,1,0,0
4096,5,75,0,0,1,0,0
4096,5,50,0,0,1,0,0
4096,5,25,0,0,1,0,0
4096,5,0,0,0,1,0,0
16384,5,100,0,0,1,0,0
16384,5,75,0,0,1,0,0
16384,5,50,0,0,1,0,0
16384,5,25,0,0,1,0,0
16384,5,0,0,0,1,0,0
32768,5,100,0,0,1,0,0
32768,5,75,0,0,1,0,0
32768,5,50,0,0,1,0,0
32768,5,25,0,0,1,0,0
32768,5,0,0,0,1,0,0
'Access specification name,default assignment
File Server,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
512,10,80,100,0,1,0,0
1024,5,80,100,0,1,0,0
2048,5,80,100,0,1,0,0
4096,60,80,100,0,1,0,0
8192,2,80,100,0,1,0,0
16384,4,80,100,0,1,0,0
32768,4,80,100,0,1,0,0
65536,10,80,100,0,1,0,0
'Access specification name,default assignment
Web Server,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
512,22,100,100,0,1,0,0
1024,15,100,100,0,1,0,0
2048,8,100,100,0,1,0,0
4096,23,100,100,0,1,0,0
8192,15,100,100,0,1,0,0
16384,2,100,100,0,1,0,0
32768,6,100,100,0,1,0,0
65536,7,100,100,0,1,0,0
131072,1,100,100,0,1,0,0
524288,1,100,100,0,1,0,0
'Access specification name,default assignment
Database - Read,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
8192,100,67,100,0,1,0,0
'Access specification name,default assignment
Database - Write,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
8192,100,33,100,0,1,0,0
'Access specification name,default assignment
Workstation - SR,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
8192,100,80,80,0,1,0,0
'Access specification name,default assignment
Workstation - X-bit labs,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
512,1,0,100,0,1,0,0
1024,2,0,100,0,1,0,0
2048,1,0,100,0,1,0,0
4096,50,60,80,0,1,0,0
8192,4,50,100,0,1,0,0
16384,6,50,100,0,1,0,0
20480,2,50,100,0,1,0,0
24576,2,50,100,0,1,0,0
28672,1,50,100,0,1,0,0
32768,13,70,70,0,1,0,0
49152,1,50,100,0,1,0,0
53248,1,50,100,0,1,0,0
65536,14,80,60,0,1,0,0
66048,2,50,100,0,1,0,0
'Access specification name,default assignment
Streaming Writes - THG,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
65536,34,100,0,0,1,0,0
131072,33,100,0,0,1,0,0
262144,33,100,0,0,1,0,0
'Access specification name,default assignment
Streaming Reads - THG,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
65536,34,0,0,0,1,0,0
131072,33,0,0,0,1,0,0
262144,33,0,0,0,1,0,0
'END access specifications
Version 2006.07.27
Scrotos wrote:So number of workers... what do people tend to use? I've seen someone say the more the better, 10 as a minimum. You indicate 1 per CPU or CPU core. I got 10 working right now on a quad core CPU and it's barely 10% CPU usage.
Scrotos wrote:Sorry if I came across that way. It was not my intention. I actually think the patterns are the easy part, since they are documented on the web. (see below)Thanks for the reprimand that I thought I'd already addressed. I do find it odd, though, since you seem to have spent quite some time trying to recreate TR's testing methodology yourself. I'd have thought you'd be more like, "yeah, post that biznatch!" rather than "RTFM."
Scrotos wrote:Basically it seemed like once Intel decided to make IOMeter open source, the distributions stopped including what Intel included before in terms of the profiles. It was unfortunate. Good thing the patterns were relatively easy to recreate. And subsequent analysis by SR debunked some patterns as not that useful anyway.I do wholeheartedly thank you for the work you did in trying to recreate the test patterns that TR uses and especially for digging up work from 3 years prior and making it publicly available! The currently available version of IOMeter has entries like:
16K; 50% Read; 0% random
16K; 25% Read; 0% random
16K; 0% Read; 0% random
...which don't seem to map to the entries you had like:
Database - Read
Database - Write
I wouldn't even know where to start looking to try and dig up various profiles that are publically available going back 6 years or longer
Scrotos wrote:even with the links you provided it seemed like a ton of reading for something I was hoping would be a quick "run this profile and find your IOps" deal. Heck, you even spent a decent amount of time researching all this and you don't even know how TR gets their concurrent requests numbers.
Dissonance wrote:All that worker, concurrent I/O, test length stuff, IMO, are the "biznatch" in terms of the setup. The clunky UI does not help things either. There was no documentation and so I just picked my own settings. I can never really compare with TR's numbers (not to mention I have different CPU+mobo). So I only made comparisons based on my own baselines. That was when I realize that I don't really need TR's exact numbers, just the trends that I needed to match.Feel free to shoot me an email if you'd like a copy of our IOMeter test file. The thing is, that file just defines the access patterns (request size, read/write distribution, randomness, etc...). You still have to configure the workers, concurrent I/O scaling, and test length manually.
Dissonance wrote:Before any tests are run, IOMeter creates a single file that spans the entire capacity of the drive, effectively putting it into a used state.
Flying Fox wrote:But yes, if any of us can post a quick "how to use IOMeter" guide, I am all for it. The strength in TR is not just the articles, but the forum community as well. Let's crowdsource this!
'Access specification name,default assignment
File Server,DISK
'size,% of size,% reads,% random,delay,burst,align,reply
512,10,80,100,0,1,4096,0
1024,5,80,100,0,1,0,0
2048,5,80,100,0,1,0,0
4096,60,80,100,0,1,0,0
8192,2,80,100,0,1,0,0
16384,4,80,100,0,1,0,0
32768,4,80,100,0,1,0,0
65536,10,80,100,0,1,0,0
'Access specification name,default assignment
File Server,ALL
'size,% of size,% reads,% random,delay,burst,align,reply
512,10,80,100,0,1,4096,0
1024,5,80,100,0,1,4096,0
2048,5,80,100,0,1,4096,0
4096,60,80,100,0,1,4096,0
8192,2,80,100,0,1,4096,0
16384,4,80,100,0,1,4096,0
32768,4,80,100,0,1,4096,0
65536,10,80,100,0,1,4096,0
We run all our tests at least three times and report the median of the results. We've found IOMeter performance can fall off with SSDs after the first couple of runs, so we use five runs for solid-state drives and throw out the first two.