WinAFC: A new affinity changer

[coo-coo-clocker]

WinAFC is a configurable program for controlling CPU affinities and priority settings of Windows applications on multi-core processor-based systems (both AMD and Intel). It is freeware available from SourceForge.net and currently runs on Vista, WinXP, and Win2K,in both 32 and 64-bit versions.

Introduction
The program is highly configurable and allows the user to control CPU affinity (including core pairing), and process priority for any application. Inspired by Folding@Home, the program is well-suited to getting the most out of complex, multi-client F@H installations.

I have been running tests with WinAFC over the last week or so to determine if it can improve F@H client performance. The results are favorable; dual-SMP output is improved using WinAFC in early testing.

Being that the program will allow detailed control over processes, it can be rather complex to fully understand and configure depending on your installation. However, the author has suggested some fairly simple rules that work great on dual-SMP/dual-GPU quad-core rigs. And best of all, with a bit of testing and analysis, we can use WinAFC to really fine-tune the performance of F@H installations. This will be increasingly important with the upcoming rollout of Intel i7 and other multi-core processors that boast 6+ cores!

Installation
Installation of WinAFC is easy. You simply unzip it into a folder of your choice, configure, and run. It does not update Windows registry, access your network, or manipulate any files outside of a few config and log files in the WinAFC installation folder. For Vista, you can run it with administrator privileges to allow it to control processes running under admin or other users.

The program reads an input file to get the configuration information it needs to control the processes of your choice. There is a help file that covers the program's configuration parameters in great detail, along with several useful examples to illustrate the many flexible features. One real advantage this program has over others of the same variety is the ability to detect which CPU cores are on the same die (in the case of Intel processors) which have shared resources (like L2 cache). This allows you to define configurations to run processes across those "pairs" to optimize their performance.

Once running, it creates a console instance (you can choose to hide this or run in "stealth mode") that provides log-style status information of the activities of WinAFC. It has a user-configurable log verbosity setting to control the level of detail of messages within the log. A nice feature is a "test mode" that you can use to see how WinAFC would alter process attributes (processor affinities, priority, etc) without actually doing it. This feature is also very useful to quickly see how the program will treat different rules you devise.

Configuration
Using the numerous parameters available in WinAFC, you can create simple or very complex rules (or "application profiles" as the author calls them). You can use process metrics such as memory or CPU utilization as input to your rules. WinAFC has several "policies" available to help you build rules that control process behavior without a lot of complex, explicit rules.

I worked with the program author quite a bit to resolve a bug, and to gain a better understanding of the program and it's configuration. He was very helpful and provided me with the configuration rules and CPU masks used in my testing. Attached to this post is the WinAFC config file used for my testing. Feel free to use it to conduct your own testing. I will not attempt to explain the various configuration parameters for WinAFC. Some of them are pretty straight-forward. Others can be quite complex, particularly when used in combination and with multiple processes in your list.

WinAFC can be very useful, since it is so highly configurable. As I mentioned above, a key advantage is being able to put like processes on the same CPU die (or "PAIR" as the author calls them). Another advantage is being able to really fine-tune each process in the F@H rig, and see what the program is doing in real-time. These features should allow us to really take affinity changing to an Xtreme level!

Testing
I ran many test configurations over the course of the last week or so. I am still testing. This will likely continue for quite some time as I play with different scenarios to further optimize F@H on my folding rig. Here is the configuration of that box:
Intel Q6600 3.3GHz, Asus P5KC, 2GB RAM, Vista 64 SP1, EVGA 9800GX2
For F@H clients, I run all beta software:
2 instances of Windows SMP 6.22b2, and 2 instances of GPU 6.20.r1

After trying several different test configs, I found two scenarios were most likely to do as well as my current affinity changer software (FAH SMP AC). So I set up the test rig to run both SMP clients on the same F@H project, and ran them approximately equal number of frames using FAH AC, WinAFC config #1, and WinAFC config #2.

Test Case #1 is FAH SMP Affinity Changer v1.0.5.0. As most of you know, the current version of FAH AC controls only SMP clients, nothing else, and really is not configurable at all, AFAIK. The docs (what little there are) are not available in english, so I really have no idea what the program really does.

Test Case #2 is WinAFC config#1. This profile does the following: put all PIDs from an SMP instance on the two CPUs that share a die ("PAIR") within the Q6600. Put the most cpu-intensive SMP PID and the least cpu-intensive SMP PID together. Put the second and third (middle two) busiest PIDs on the other CPU of the pair. Then put one GPU client with the two middle SMP PIDs. Do everything like this again for the second instance of SMP and GPU onto the other pair of CPUS.

Test Case #3 is WinAFC config#2. This profile does the following: Same setup as config #1, but allow the least expensive SMP PIDs to roam on both cores of the pair, based on how busy the core is.

I'll jump straight to the results first, then do some additional explaining:



Looking at the bottom of the chart, you can see the PPD results from the 3 test scenarios. The first scenario (FAH AC) is the baseline, with 13,864 PPD for all four F@H clients. The second scenario is the WinAFC Config #2, with 13,802, performing just a bit under the FAH AC. The final scenario is running WinAFC config #1, scoring marginally better than the baseline with 14,112 total PPD.

How to interpret the details:
Each process (PID) of the F@H client is a row within each group. The numbers along that row are the CPU affinity assigned to that PID during the test. Avg Frame is the mean average duration of a frame during that test case. The memory and CPU time numbers down the right side are the resource utilization captured near the end of each project for each PID. For GPU clients, this number is an average across the projects during the test period.

During the testing, I used Vista Task Manager to identify PIDs, analyze where they were running, their priority, utilization, etc. Here's a shot during execution of the test:


Summary
WinAFC is highly configurable and provides great activity reporting. You can control every process in your system, not just isolated ones. Early test results show only marginal improvements over FAH SMP AC. However, I have really only scratched the surface on this. Given more time to experiment, I am optimistic I can increase F@H client performance well beyond the marginal gains I have acheived in my preliminary testing.

[catawalks]

Long, but looks good so far.

Nice work.

[Marvin_The_Martian]

Great work

Abit dissapointed seeing the gains, I had hoped for more. I was posting about it in another thread not even seeing you had already posted this. Think I will do some testing as well with my vm's, try if pairing those on one die will hold a significant point boost or not.

[-pWs-]

Here is my setting.

[Marvin_The_Martian]

-pWs- you got some results for us as well? Ppd comparisons of using it vs blanco ( let the os manage the threads itself )? That would certainly add to the knowledge pool and give a clearer picture about the usefullness

[-pWs-]

Okie.

C:\Folding@Home\Windows SMP - 1\FahCore_*.exe := PAIR0::CPU0+PAIR1 [assign=1,resource=MEMUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP - 2\FahCore_*.exe := PAIR0::CPU1+PAIR1 [assign=1,resource=MEMUSE,policy=PSEUDOBALANCED]

C:\Folding@Home\Nv GPU\FahCore_*.exe := PAIR0 [assign=1,resource=MEMUSE,policy=PSEUDOBALANCED]

Results:
SMP 1: 18m37s , SMP 2: 17m32s Total: 36m9s

C:\Folding@Home\Windows SMP - 1\FahCore_*.exe := PAIR0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP - 2\FahCore_*.exe := PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

Results:
SMP 1: 15m45s , SMP 2: 16m38s Total: 32m23s

C:\Folding@Home\Windows SMP - 1\FahCore_*.exe := PAIR0::CPU0+PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP - 2\FahCore_*.exe := PAIR0::CPU1+PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

C:\Folding@Home\Nv GPU\FahCore_*.exe := PAIR0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

Results:
SMP 1: 19m , SMP 2: 16m59s Total: 35m59s

C:\Folding@Home\Windows SMP - 1\FahCore_*.exe := CPU0+CPU1+CPU2+CPU3 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP - 2\FahCore_*.exe := CPU3+CPU2+CPU1+CPU0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

Results:
SMP 1: 22m15s , SMP 2: 22m14s Total: 44m29s

C:\Folding@Home\Windows SMP - 1\FahCore_*.exe := CPU0+CPU1+CPU1+CPU0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP - 2\FahCore_*.exe := CPU2+CPU3+CPU3+CPU2 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

Results:
SMP 1: 18m53s , SMP 2: 16m39s Total: 35m32s

[Marvin_The_Martian]

you had that and didn't post it in the first post lol

In the results with no gpu policy, you're not running the client at all or just not managing it's thread affinity?

[-pWs-]

All settings are w Dual SMP + GPU2.
Yup. Not managing its thread affinity.

[p2501]

Ok, I downloaded WinAFC64 and it dies on startup (with and without Admin elevation), nothing else happens. Did you guys try the 32bit one?

I really want to try a new affinity manager, because this process lasso thingie isn't doing what it should. It is not considering different folders with the same named exetutable.

[Marvin_The_Martian]

Ok, I downloaded WinAFC64 and it dies on startup (with and without Admin elevation), nothing else happens. Did you guys try the 32bit one?

I really want to try a new affinity manager, because this process lasso thingie isn't doing what it should. It is not considering different folders with the same named exetutable.

I think it would help if you posted your config file? And if I where you I would ask Dak1640 about ProcessLassoo

[p2501]

I think it would help if you posted your config file? And if I where you I would ask Dak1640 about ProcessLassoo

I didn't configure anything beforehand, so I guess the problem lies with me. But reading the help file, I found something very useful:

*\vmware-vmx.exe := PAIR0+PAIR1 [assign=2,priority=idle]

Awesome tool, I hope with the config file it'll work, this is exactly what I wanted. Btw Marvin, you where right about deleting the VM's and start anew, it's chugging along now. Let's see if it stays running this time.

[coo-coo-clocker]

Ok, I downloaded WinAFC64 and it dies on startup.

what kind of error are you getting? Which OS are you running?

[p2501]

what kind of error are you getting? Which OS are you running?

I run Vista64 Ultimate, UAC enabled. I just dies without any kind of error output, not even when I run it in a console. I now configured it, but the same thing happens:

#
# Affinity Changer Console Input File
#
# Lines that start with a '#' and empty lines are ignored. 
# Configuration lines include global flag setting lines, and application profile
# lines. These lines can be interleaved in any order. However, the order in which
# you specify the various application profiles matters if a process could be
# matched by multiple application profiles.

# To run AffinityChanger in TestMode (no affinity settings are modified),
# uncomment the line below and specify a value different than zero. To disable 
# test mode, set the value to 0 or comment the line below.
# It is recommended that you try the input file in test mode first, after you
# make changes, especially if you are unfamiliar with the syntax.
TestMode = 1

# There are several flags that control the verbosity of the program. In general,
# a higher value means a more verbose output. 

# By default the warning level is set to 2. To change it, uncomment the next 
# line and specify a value>=0. Zero disables warning messages (not recommended).
#WarningLevel = 2

# Logging is done on both console and in a file called 'affinitylog.txt'. 
# Separate flags control the verbosity for each medium. By default the logging
# level is set to 3. Uncomment the next lines to change the values of these 
# flags.
# To change only the console verbosity:
ConsoleLogLevel = 4
#
# To change only the file verbosity:
#FileLogLevel = 3
#
# To change both mediums at the same time:
#LogLevel = 3

# By default AffinityChanger checks for new processes every 30 seconds.
# To change how frequently AffinityChanger checks for new processes or adjusts
# affinities for existing processes, uncomment the line below and specify the
# number of seconds between consecutve process checks. The minimum accepted 
# value is 1 and the value can be only an integer number.
#TimeInterval = 30

# One can change the value of a flag multiple times. In these cases the 
# most recently specified value will take effect.
#

##################################################
##          Application Profile lines
##################################################
#
# An application profile is specified on a single line. 
# An application profile includes the following information: an application
# name, a CPU mask, and optional attributes in the following format:
# C:\Path\To\Application := CPU0+CPU1 [attr1=val1,attr2=val2]
#
# Check the documentation and the provided examples for more 
# information about these fields.
C:\Program/ /Files/ /(x86)\VMware\VMware/ /Player\x64\vmware-vmx.exe := PAIR0+PAIR1 [assign=2,priority=idle]
C:\Users\root\FAH-GPU\FahCore_11.exe := PAIR0::CPU1 [priority=High]
C:\Users\root\FAH-GPU2\FahCore_11.exe := PAIR1::CPU1 [priority=High]

I really want to use this, what am I doing wrong?

[Sparky]

Hmm would this be beneficial running 1 SMP and 1 ATI GPU?

[p2501]

Hmm would this be beneficial running 1 SMP and 1 ATI GPU?

I think so, Sparky. Just download it and read the documentation, there is something about sorting SMP threads. You could try it and supply us with some benches...

Edit: For 64bit VMWare users, in the official version there is a bug (or better, an optimization too much ), which leads to it to die on startup. Here (last post) is a fixed version, works like a charm.

[coo-coo-clocker]

let me know if that new version is still causing you grief...
if it is still crashing on start, please open up the error detail box and screen shot that. Thanks

[p2501]

let me know if that new version is still causing you grief...
if it is still crashing on start, please open up the error detail box and screen shot that. Thanks

There is no error detail box, it just died with Windows telling me to close it, no error details (see post #13). I already talked about it with baobab (in the thread I linked) and the error for 64bit was already known (even though he knew only one more person with the same problems).

So, if anyone is experiencing problems with the 64bit WinAFC, please go here:

http://foldingforum.org/viewtopic.ph...&sd=a&start=60

The last post has a fix. Also, if you don't have problems on 64bit Vista with the original from sourceforge, go there anyway so baobab can find out details. Well, he knew what was wrong already, so just go there and read up.

Thanks coo-coo-clocker, without you telling us I wouldn't have this nice little program running here. It does exactly (setting just affinities and priorities) what I want and just that.

[Sparky]

I downloaded, ran, and was totally clueless what to do lol. It is late - or early, depending on how you look at it - so my brain isn't in full power mode anyway.

[p2501]

I downloaded, ran, and was totally clueless what to do lol. It is late - or early, depending on how you look at it - so my brain isn't in full power mode anyway.

This is what I did after downloading it the first time , too!

Well, you downloaded it, just go have a read now. You need to set it up before launching it. The readme is exhaustive, I set mine up according to that and it works works works works.......

It's nice and small. Awesome lil' proggy.

[coo-coo-clocker]

at the bottom of the first post I attached a copy of my config file.
Take a look at that. Perhaps once peeps with other OS and configs get it tweaked and running well, then they can post their config files here too and it will be plug and play.

[Marvin_The_Martian]

I downloaded it to, but want some more datapoints without the app for comparison, also just about to cycle vm wu's and I think I should do a compare either per full wu or atleast 10 frames between 20% and 80%. I'm going to try and help but I'm spreading my limited resouces ( = capacity ) thin already

[OldChap]

Okie.

C:\Folding@Home\Windows SMP - 1\FahCore_*.exe := PAIR0::CPU0+PAIR1 [assign=1,resource=MEMUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP - 2\FahCore_*.exe := PAIR0::CPU1+PAIR1 [assign=1,resource=MEMUSE,policy=PSEUDOBALANCED]

C:\Folding@Home\Nv GPU\FahCore_*.exe := PAIR0 [assign=1,resource=MEMUSE,policy=PSEUDOBALANCED]

Results:
SMP 1: 18m37s , SMP 2: 17m32s Total: 36m9s

C:\Folding@Home\Windows SMP - 1\FahCore_*.exe := PAIR0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP - 2\FahCore_*.exe := PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

Results:
SMP 1: 15m45s , SMP 2: 16m38s Total: 32m23s

C:\Folding@Home\Windows SMP - 1\FahCore_*.exe := PAIR0::CPU0+PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP - 2\FahCore_*.exe := PAIR0::CPU1+PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

C:\Folding@Home\Nv GPU\FahCore_*.exe := PAIR0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

Results:
SMP 1: 19m , SMP 2: 16m59s Total: 35m59s

C:\Folding@Home\Windows SMP - 1\FahCore_*.exe := CPU0+CPU1+CPU2+CPU3 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP - 2\FahCore_*.exe := CPU3+CPU2+CPU1+CPU0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

Results:
SMP 1: 22m15s , SMP 2: 22m14s Total: 44m29s

C:\Folding@Home\Windows SMP - 1\FahCore_*.exe := CPU0+CPU1+CPU1+CPU0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP - 2\FahCore_*.exe := CPU2+CPU3+CPU3+CPU2 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

Results:
SMP 1: 18m53s , SMP 2: 16m39s Total: 35m32s

I've been playing around with this some and can't get anywhere near these times...maybe because I'm running 3*GPU, maybe because of the 57xx wu's
Anyone want to share settings hints or tips?

Edit: q6600@3.6 4gig@1066 xp sp3

[coo-coo-clocker]

I've been playing around with this some and can't get anywhere near these times...maybe because I'm running 3*GPU, maybe because of the 57xx wu's
Anyone want to share settings hints or tips?

Edit: q6600@3.6 4gig@1066 xp sp3

XP is your problem there, OC.
You'll need to put Vista on in order to reduce the CPU load from the GPU clients.

[MikeB12]

you might want to try that new beta xp nvidia drivers, it "supposedly" removes the xp fahcore_11 cpu core hawg issue.
read thread first, I'm vista, so I'm not gonna read all umpteen pgs.
http://foldingforum.org/viewtopic.php?f=52&t=7063

[OldChap]

Thanks c-c-c, I hadn't thought of that one
Waiting on a maximus rma, got everything else including 45nm, water, chiller and vista for that one meanwhile I'm going to be the guinea pig and give 180.60 a go just to see how that goes. want to guess? higher or lower?

Edit: Mike....you saw some of that info too huh? well here's hoping