WinAFC: A new affinity changer

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:

http://mysite.verizon.net/vzewvvxs/s...inity_test.jpg

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:
http://mysite.verizon.net/vzewvvxs/s...s/fah_pids.jpg

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.

Long, but looks good so far.:up:

Nice work.

Great work :up:

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.

Here is my setting. :up::up:

-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 :up:

Okie. :up::up:

Code:

---

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

Code:

---

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

Code:

---

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

Code:

---

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

Code:

---

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

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

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

:D:D

All settings are w Dual SMP + GPU2. :up::cool:
Yup. Not managing its thread affinity. :)

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.

Quote:

---

Originally Posted by 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.

---

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

Quote:

---

Originally Posted by Marvin_The_Martian

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.

Quote:

---

Originally Posted by p2501

Ok, I downloaded WinAFC64 and it dies on startup.

---

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

Quote:

---

Originally Posted by coo-coo-clocker

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:

Code:

---

#
# 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]

---

http://bandwidth.se/imgs/101/0/winafcerror.png

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

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

Quote:

---

Originally Posted by SparkyJJO

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... :p:

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. :)

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

Quote:

---

Originally Posted by 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

---

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. :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. :yepp:

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

Quote:

---

Originally Posted by SparkyJJO

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

---

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

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. :up:

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.

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 :rofl: ) thin already :up:

Quote:

---

Originally Posted by -pWs-

Okie. :up::up:

Code:

---

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

Code:

---

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

Code:

---

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

Code:

---

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

Code:

---

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

Quote:

---

Originally Posted by OldChap

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.

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

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

Hehe I was just reading up again in that thread and the things I read are conflicting. One says higher temps, lower cpu usage, other says no change besides ppd drop...

I'm not on xp anyway but jeez it's messy info :shrug:

Yep sure is a mess same goes for the ability to use rivatuner although I think the latest version will do it.

Well if all goes to plan by sunday night we could have a few answers even if only for my cards.

You know it's just people either don't post enough info or they post conflicting info, I rather get info from someone I know and who will give all the specs included and maybe even some ss's to back him up.Not that I don't trust people, I turst the person but he might have done something wrong or forgot an important side issue :poke:

:sofa:

Honestly, I think many don't have the aptitude (or perhaps the attention span) to do anything close to useful testing. If it ain't plug and play, they don't mess with it... :rolleyes:

I grew up in the days where you really had to know some stuff about computers to be a user. Not saying that it should be that way now - I mean it's nice to not have to know so much, but it sure does help to at least have a fundamental understanding of the hardware and software.

I don't pretend to have the knowledge you guys have. For me it's just a case of noting the performance of various combinations of wu's and affinities with current drivers then switch to 180.60 upgrade rivatuner and start over. Lastly I'll go over my temp chart and see if that has changed.

Frankly because I've only recently oc'd to 3.6 and added smp *2 My info on this part of the equasion is less extensive, that said, a better or worse conclusion should be possible.

If anyone has a particular Idea or request before I start on this (in about 4 hours) let me know.

Cool! Let us know how it goes OC.
Since I don't have a quad running XP, I'm interested in seeing the results. It could help many here. Here are the things I'd like to see tried out:

1) Pairing each SMP instance on one die (share an L2 cache):

Code:

---

C:\Folding@Home\Windows SMP1\FahCore_*.exe := PAIR0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP2\FahCore_*.exe := PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

---

3) Now sprinkle in the GPU clients (one client per pair)

Code:

---

C:\Folding@Home\Windows SMP1\FahCore_*.exe := PAIR0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP2\FahCore_*.exe := PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
*\Folding@home-gpu1\FahCore_*.exe := PAIR0 [force=yes,priority=belownormal]
*\Folding@home-gpu2\FahCore_*.exe := PAIR1 [force=yes,priority=belownormal]

---

2) Distributing the busiest PIDs from all clients across all cores:

Code:

---

*\FahCore_*.exe := ALL [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

---

Hmmm this is tricky without testing it. Not sure if I have all of this correct :D

Quote:

---

Originally Posted by coo-coo-clocker

Cool! Let us know how it goes OC.
Since I don't have a quad running XP, I'm interested in seeing the results. It could help many here. Here are the things I'd like to see tried out:

1) Pairing each SMP instance on one die (share an L2 cache):

Code:

---

C:\Folding@Home\Windows SMP1\FahCore_*.exe := PAIR0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP2\FahCore_*.exe := PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

---

3) Now sprinkle in the GPU clients (one client per pair)

Code:

---

C:\Folding@Home\Windows SMP1\FahCore_*.exe := PAIR0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
C:\Folding@Home\Windows SMP2\FahCore_*.exe := PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]
*\Folding@home-gpu1\FahCore_*.exe := PAIR0 [force=yes,priority=belownormal]
*\Folding@home-gpu2\FahCore_*.exe := PAIR1 [force=yes,priority=belownormal]

---

2) Distributing the busiest PIDs from all clients across all cores:

Code:

---

*\FahCore_*.exe := ALL [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED]

---

Hmmm this is tricky without testing it. Not sure if I have all of this correct :D

---

That looks way too complicated for a lowly Windows guy. :( I just loves me some easy GUIs!!!

http://img355.imageshack.us/img355/8...nityguiub6.png
http://img355.imageshack.us/img355/p...png/1/w799.png

Just an interim update
The job has been made somewhat harder by having different points on GPU now but at first glance (all 511pts GPU) there seems to be little or no difference how the affinities are set. I'll look again after a nights sleep.

I had to check if GPU was running with these drivers (180.60) there is minimal cpu use and no major ppd loss but more on this when I have more data

WFO it is an open source project under gpl, if you guys put up enough data I think it's not that hard to make a simple gui with common profiles, when these a'holes here who are supposed to be pulling me out of hole instead of kicking me deeper down in it finally learn about that and hopefully start doing just that I'll have more time and I'll look into it :up:

Wierd issues here I guess I need to switch this off until I see what's happening with the 180.60's ...went to bed with fahmon showing the smp's running @ 1069 and 1069 woke up and now showing 845 and 826 both readings show no change to GPU and are after min 3%

Ugh. Can you post the WinAFC affinitylog.txt?

I've stopped one smp for now I'll finish these two up before trying this again. A fresh start so to speak.

From using priaffinity I have a hunch that letting everything roam is best as to the low or below normal I'll have to wait see

Code:

---

Auto detecting CPU configuration. Please wait.
29/11/2008 19:00:03: Found 4 CPUs:
        CPU0 has mask 00000001
        CPU1 has mask 00000002
        CPU2 has mask 00000004
        CPU3 has mask 00000008
29/11/2008 19:00:03: Found 2 PAIRs:
        PAIR0<CPU0,CPU1> has mask 00000003
        PAIR1<CPU2,CPU3> has mask 0000000C
29/11/2008 19:00:03: C:\FAH\FAH 622.exe with PID 3616 using  0.1% of CPU time, set affinity to CPU0
29/11/2008 19:00:03: C:\FAH\FAH 622.exe with PID 3616, set priority to Idle
29/11/2008 19:00:03: C:\FAH2\FAH 622.exe with PID 3976 using  0.0% of CPU time, set affinity to CPU2
29/11/2008 19:00:03: C:\FAH2\FAH 622.exe with PID 3976, set priority to Idle
29/11/2008 20:39:18: C:\FAH\FAH 622.exe with PID 3616 using  0.0% of CPU time, set affinity to CPU0
29/11/2008 20:39:18: C:\FAH\FAH 622.exe with PID 3616, set priority to Idle
29/11/2008 20:39:18: C:\FAH2\FAH 622.exe with PID 3976 using  0.0% of CPU time, set affinity to CPU2
29/11/2008 20:39:18: C:\FAH2\FAH 622.exe with PID 3976, set priority to Idle

---

This was my first try

Using priaffinity (which I understand better) the figures suggest that the best setup is to let smp1 roam on cores 0 and 1, let smp2 roam on cores 2 and 3, then let any instances of GPU roam freely across all 4 cores. the core pairings match pair0 and pair1 in above post

If you understand the winafc settings and could suggest a set to match this scenario I would be grateful (my brain is totally jailed just now) otherwise I'll try to get this later (probably tomorrow)

ok, try this:

Code:

---

C:\F@H SMP1\FahCore_*.exe := PAIR0 [assign=2,resource=CPUUSE,policy=PSEUDOBALANCED,priority=idle]
C:\F@H SMP2\FahCore_*.exe := PAIR1 [assign=2,resource=CPUUSE,policy=PSEUDOBALANCED,priority=idle]
C:\F@H gpu*\FahCore_*.exe := ALL [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED,priority=belownormal]

---

I'm heading out to work now, but I can try this out later or let me know if it does what you like...

Thanks CCC, I set it up in a while and compare the figures....I'm now using 180.60 so the effects should be clear.

Is it possible to utilise this without some of the arguements being defined?

At first glance (by a software inept old guy with poor memory) for smp all I want this to do is select pair0 or pair1, the pseudobalancing is perhaps unnecessary.

My initial results (after 5%) are of ppd rates of 1287 and 1457 this is with a new wu on both smp's so that is a variable. My previous trials (priaffinity) have shown ppd's as below

1665 1640
1662 1607
1724 1495
1673 1527

These figures depend on the combination of GPU wu's

Edit: ok I read the instructions again and am currently starting again with just cores specified. I will add other perameters one at a time with this.

Well, I'm no expert either, I've just been playing with it for a while.
The balancing parms I thought would provide the "roaming across the pair" action that you wanted...rather than just pegging it to one of the CPUs in the pair.

Don't forget to take the "test" mode flag out - or it won't apply any of the changes.

Let me monkey with some testing and see what I get.

Let me clarify a few things since I wrote that program.

First, coo-coo-clocker pointed me to this thread when he posted the initial review. We had a bit of a discussion through email when he was trying to configure the program for his systems. I registered on the forum at that time because I noticed one or two messages that needed an answer. But, I did not receive the email with the activation link until today. So, today I checked the thread again and I see a few other posts to which I might be able to give an answer.

Quote:

---

Is it possible to utilise this without some of the arguements being defined?

---

Yes. If you want to use the same mask for all processes matched by a particular application profile, then you do not need to specify any optional attributes (those things between []). Thus, if you want to run one SMP per PAIR with all four processes roaming on both cores of the PAIR, it is sufficient to say:

Code:

---

C:\FAH\FahCore_*.exe := PAIR0
C:\FAH2\FahCore_*.exe := PAIR1

---

You do not have to specify any assign, resource or policy attributes in this case.
You are free to specify the priority attribute if you want to change the priority of those processes.

Quote:

---

Code:

Code:

---

Auto detecting CPU configuration. Please wait.
29/11/2008 19:00:03: Found 4 CPUs:
        CPU0 has mask 00000001
        CPU1 has mask 00000002
        CPU2 has mask 00000004
        CPU3 has mask 00000008
29/11/2008 19:00:03: Found 2 PAIRs:
        PAIR0<CPU0,CPU1> has mask 00000003
        PAIR1<CPU2,CPU3> has mask 0000000C
29/11/2008 19:00:03: C:\FAH\FAH 622.exe with PID 3616 using  0.1% of CPU time, set affinity to CPU0
29/11/2008 19:00:03: C:\FAH\FAH 622.exe with PID 3616, set priority to Idle
29/11/2008 19:00:03: C:\FAH2\FAH 622.exe with PID 3976 using  0.0% of CPU time, set affinity to CPU2
29/11/2008 19:00:03: C:\FAH2\FAH 622.exe with PID 3976, set priority to Idle
29/11/2008 20:39:18: C:\FAH\FAH 622.exe with PID 3616 using  0.0% of CPU time, set affinity to CPU0
29/11/2008 20:39:18: C:\FAH\FAH 622.exe with PID 3616, set priority to Idle
29/11/2008 20:39:18: C:\FAH2\FAH 622.exe with PID 3976 using  0.0% of CPU time, set affinity to CPU2
29/11/2008 20:39:18: C:\FAH2\FAH 622.exe with PID 3976, set priority to Idle

---

This was my first try

---

What is FAH_622? I assume this is the folding client, not the FahCore. It uses < 0.1% of a CPU. You should have set the affinity of the FahCore_*.exe processes.

Quote:

---

That looks way too complicated for a lowly Windows guy. I just loves me some easy GUIs!!!

---

I do not disagree with you. There is a trade-off between flexibility and ease of use. If the program you are using does everything you need, there is no need to try something more complicated.
All these programs do a very simple thing in the end, something that you could do by hand. It is just a convenience and there should be no difference in the performance of the targeted applications if you set the same affinities, just using different tools.

I popped-in just to clarify a few things when my account finally got created. If you need answers to other questions, and you cannot get them here, you can always ask a question in the thread created at the folding forum, since I check that one more often.

Happy folding.:up:

Would you mind if I'd write a gui around it?

Might even include it in maxFah/maxTray, built in affinty setter :)

So starting again:

These numbers do not stay fixed but are constantly moving

C:\FAH\FahCore_*.exe := PAIR0 [priority=idle]
C:\FAH2\FahCore_*.exe := PAIR1 [priority=idle]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU_card 0\Fahcore_*.exe := PAIR0+PAIR1 [priority=Belownormal]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU_card 1\FahCore_*.exe := PAIR0+PAIR1 [priority=Belownormal]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU_card 2\FahCore_*.exe := PAIR0+PAIR1 [priority=belownormal]

ppd values of wu's.......... gpu2 gpu1 gpu0 smp2 smp1 total
480+480+480+1760+1940 6283 5760 5760 1852 1512 21167
480+480+511+1760+1940 6283 5760 4204 1934 1576 19759

No time till this evening to add more parameters by which time there will be a new wu. but all data is good data right?

Looking better!!
I think you can also specify "ALL" as equivalent for PAIR0+PAIR1

Quote:

---

Originally Posted by Marvin_The_Martian

Would you mind if I'd write a gui around it?

Might even include it in maxFah/maxTray, built in affinty setter :)

---

I do not mind. If you can write a GUI that is intuitive and enables a user to specify all the attributes, that would be very useful.

Quote:

---

Originally Posted by OldChap

So starting again:

These numbers do not stay fixed but are constantly moving

C:\FAH\FahCore_*.exe := PAIR0 [priority=idle]
C:\FAH2\FahCore_*.exe := PAIR1 [priority=idle]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU_card 0\Fahcore_*.exe := PAIR0+PAIR1 [priority=Belownormal]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU_card 1\FahCore_*.exe := PAIR0+PAIR1 [priority=Belownormal]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU_card 2\FahCore_*.exe := PAIR0+PAIR1 [priority=belownormal]

ppd values of wu's.......... gpu2 gpu1 gpu0 smp2 smp1 total
480+480+480+1760+1940 6283 5760 5760 1852 1512 21167
480+480+511+1760+1940 6283 5760 4204 1934 1576 19759

No time till this evening to add more parameters by which time there will be a new wu. but all data is good data right?

---

I'm not going to comment about the numbers moving. There will always be a slight fluctuation, especially if you also use your machine for other light tasks.

Also, I am not going to comment the optimality of the config. There are many possible configurations. Let me just say that you could replace the three GPU profiles with a single one that takes care of all three GPU clients, like so:

Code:

---

C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU*\Fahcore_*.exe := ALL [priority=belownormal]

---

This is just to show that if you want to do simple stuff, the rules are simple. They become more complicated when you try to do some balancing, or assign a different number of cores to each process.

Quote:

---

Originally Posted by coo-coo-clocker

Looking better!!
I think you can also specify "ALL" as equivalent for PAIR0+PAIR1

---

That is correct. ALL is a shortcut for all CPUs in a machine.

Thanks for your input on this. Actually the more I "play around" with this the easier it is getting. Printing and reading the guide again helped too. :rolleyes:

For the record it is already doing what I wanted from this. Further tweaking is really just to answer a few questions about Fah.

One quick question though ...

If I want to remove all previous settings before starting another test run am I ok with just leaving the perameter blank or must I correct that setting to its default?

Thanks again for a useful tool :up::up::up:

Quote:

---

Originally Posted by OldChap

Thanks for your input on this. Actually the more I "play around" with this the easier it is getting. Printing and reading the guide again helped too. :rolleyes:

For the record it is already doing what I wanted from this. Further tweaking is really just to answer a few questions about Fah.

One quick question though ...

If I want to remove all previous settings before starting another test run am I ok with just leaving the perameter blank or must I correct that setting to its default?

---

Are you referring to the optional attributes of a profile? You do not have to restore them to a default value. In fact the program is not aware that you load new profiles for the same processes. Currently you cannot just load a new application profile on top of the existing ones. When you load a new configuration file, all the active profiles are purged from memory, and the new profiles replace them. The new profiles may match the same processes, or a different set of processes.

Attributes are specified for each profile separately. If you do not specify an attribute, a default value is considered. Attribute default values cannot be changed by the user. You can only specify the values you want to use, otherwise the program defaults are used.

However, once you set an affinity mask for a process, the mask persists even after you exit WinAFC. To change a process affinity mask, you have to set it explicitly to the new desired value. Affinity masks are communicated to the OS, in this case Windows, to be used by the OS scheduler. Windows stores the masks internally, in a per process structure. The mask will persist until either the process terminates, or until the user changes it explicitly (either manually or using a tool).

If you want to restore the affinity masks to their initial values when you exit WinAFC, or when you load a new configuration file, or when you pause WinAFC, you have to select the respective "restore affinity" options in Settings -> General Settings.

Did I confuse you enough? :rofl: I do not even know if this is what you were asking. Maybe you can rephrase the question if this is not what you wanted to know.

Nailed it!! It was the question about persistancy.....Logically after quitting WinAFC, settings would remain until the program re started or I changed them but I wanted to ask anyway.

Now some results:-

C:\FAH\FahCore_*.exe := PAIR0 [assign=2,resource=CPUUSE,policy=PSEUDOBALANCED,pri ority=idle]
C:\FAH2\FahCore_*.exe := PAIR1 [assign=2,resource=CPUUSE,policy=PSEUDOBALANCED,pri ority=idle]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU*\Fahcore_*.exe := ALL [priority=belownormal]
ppd values of wu's......... gpu2 gpu1 gpu0 smp2 smp1 total
348+511+511+1760+1940 4551 4204 5681 1912 1564 17912
384+384+480+1760+1940 6144 5623 5760 1962 1572 21061
384+384+480+1760+1940 6144 5720 5681 1967 1579 21091
480+480+480+1940+1940 6283 5760 5760 1559 1559 20921
384+384+384+1940+1940 6144 5720 5720 1556 1627 20767

Quote:

---

Originally Posted by OldChap

Now some results:-

C:\FAH\FahCore_*.exe := PAIR0 [assign=2,resource=CPUUSE,policy=PSEUDOBALANCED,pri ority=idle]
C:\FAH2\FahCore_*.exe := PAIR1 [assign=2,resource=CPUUSE,policy=PSEUDOBALANCED,pri ority=idle]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU*\Fahcore_*.exe := ALL [priority=belownormal]

---

OC, just another note to say that these rules are equivalent to the ones that you used before.

PAIR0 is a shorthand for writing PAIR0::CPU0 + PAIR0::CPU1 (2 CPUs in total).
For the sake of generality let's assume you specify the 'assign' attribute with a value of N and you have M CPUs in the CPU mask.

If N = M, you tell the program to assign all CPUs listed in the CPU mask to all matching processes. In this case the attributes 'resource' and 'policy' are ignored. If N = M, the program does not have to compute resource consumption for matching processes, does not sort the processes by said resource consumption, and does not have to allocate the cores using some fancy policy.
The program just sets the affinity mask of all matching processes to include all CPUs listed in the CPU mask.

So those rules are equivalent to just saying:

Code:

---

C:\FAH\FahCore_*.exe := PAIR0 [priority=idle]                                               
C:\FAH2\FahCore_*.exe := PAIR1 [priority=idle]                                               
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU*\Fahcore_*.exe := ALL [priority=belownormal]

---

Attributes 'resource' and 'policy' are meaningful only when N < M. Note that N cannot be greater than M.

Edit: If you want to test a different config, I would test this one:

Code:

---

C:\FAH\FahCore_*.exe := PAIR0 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED,priority=idle]
C:\FAH2\FahCore_*.exe := PAIR1 [assign=1,resource=CPUUSE,policy=PSEUDOBALANCED,priority=idle]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU*\Fahcore_*.exe := PAIR0::CPU1+PAIR1::CPU1 [priority=belownormal]

---

If your OS is XP and not Vista, then I recommend replacing resource=cpuuse with resource=memuse for both SMP profiles.

Thanks again, Yes I'm using xp so I will give these a go after work

Some more results...This time in line with baobab13's recommendations. For me This whole excercise has been a resounding success. Until I finish building another dedicated rig with linux and wine this is the best I could hope for.

My thanks to you baobab13...great piece of work!!

Feel free to ask if you want any other settings tested


C:\FAH\FahCore_*.exe := PAIR0 [assign=1,resource=MEMUSE,policy=PSEUDOBALANCED,pri ority=idle]
C:\FAH2\FahCore_*.exe := PAIR1 [assign=1,resource=MEMUSE,policy=PSEUDOBALANCED,pri ority=idle]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU_card*\Fahcore_*.exe := PAIR0::CPU1+PAIR1::CPU1 [priority=belownormal]
ppd values of wu's.......... gpu2 gpu1 gpu0 smp2 smp1 total
511+480+511+1940+1940 4599 5760 4245 1866 1902 18372
480+384+480+1940+1940 6283 5623 5760 1849 1898 21413
384+384+384+1940+1940 6259 5623 5623 1835 1915 21255
480+384+480+1760+1760 6283 5623 5760 2085 2063 21814
511+384+511+1760+1760 4599 5720 4204 2262 2080 18865

Quote:

---

Originally Posted by baobab13

I do not mind. If you can write a GUI that is intuitive and enables a user to specify all the attributes, that would be very useful.

---

I am concidering just using it as 3rd party enhanchement for my client installer/manager/monitor which I'm still working on.

The simple and intuitive is hard with the abundant options, and I was hoping getting enough profiles tested by allot of people to just be able to use predetermined profiles based on installed hw, clients and operating system. I got all three of those, I just don't know yet how to generate the best profile for them ( I am following your threads though, and there are more and more cases where it seems a predetermind profile will work ). However if some other affinity changer is used, or the user run software for which it want to create some breating room, that's where it might get tricky.

When I got the basic install validated, and the tray control working fully I'll start on including afc. I done some testings months back but I got permission problems when trying to change affinities, think I know now how to avoid them but it's abit like trying to reinvent the wheel ;) I'd rather use your excellent program and support :up:

Marvin, that's fine as well. It is an open source project. I guess the GPL license requires only that you make your tool open source as well. :)

I can also answer questions if you'll have any when you start. I have a few other features that I want to add as well (though a GUI is not on the list :) ). That's why the program is not at version 1.0 yet.
However, I am a bit more busy now till the end of the year, so these things are on hold.

Actually I don't have to afaik, I can give users the option to dl your program, and then write profiles for it to read. As I'm not using the sourcecode in my program gpl does not apply?

I'm not against open source, not at all but I spend allot of time on this, and when it's done I want to dedicate it to someone before I will release the sourcecode. I uploaded some much older snippets on google code awhile back, but I think about 60% of the code has been cleaned/improved on so it's not representative of the current state!

Quote:

---

Originally Posted by Marvin_The_Martian

Actually I don't have to afaik, I can give users the option to dl your program, and then write profiles for it to read. As I'm not using the sourcecode in my program gpl does not apply?

I'm not against open source, not at all but I spend allot of time on this, and when it's done I want to dedicate it to someone before I will release the sourcecode. I uploaded some much older snippets on google code awhile back, but I think about 60% of the code has been cleaned/improved on so it's not representative of the current state!

---

I misunderstood. I thought you wanted to incorporate the code in your program and create/modify profiles dynamically through a GUI. Now I see that you want a GUI to generate the file. Sure, the GPL does not apply in this case.
I am not trying to force open source on you, but in general, when I have a choice I prefer an open source code over a closed one (to download).
Cheers.

That's true and I want to open source it but I want to finish it before I do. I know my code can be improved upon greatly and I am not doing it for personal glory but I am doing it because I lost people to cancer and this is the only way I could be sure that's rememberd. If I was less afraid of being overlooked, I would opensource it right now, infact I have google code pages where older versions are on, but I haven't put the new config form on it as it's the largest piece of code of the whole program, and also a 100% working way to wrap a gui around the console clients.

Think I just been dissapointed in people to much in the past to have enough trust in the open source community to give me some credit for the time I spend on this :(

When I get some more confirmation the installer works, that part will be opensourced next week, and I'll work on the tray control for the cllients and the log parsing/qeueu.dat parsing/client control next. I want to include the option to install clients on remote windows computers in the traycontrol, so you can set up a windows farm from a centralized application. I know how to do it, I know it will take some trail and error but I think I can manage it.

I'm all for open source :yepp: I just want to show something off which works reliable first :sofa:

Quote:

---

Originally Posted by OldChap

Some more results...This time in line with baobab13's recommendations. For me This whole excercise has been a resounding success. Until I finish building another dedicated rig with linux and wine this is the best I could hope for.

My thanks to you baobab13...great piece of work!!

Feel free to ask if you want any other settings tested


C:\FAH\FahCore_*.exe := PAIR0 [assign=1,resource=MEMUSE,policy=PSEUDOBALANCED,pri ority=idle]
C:\FAH2\FahCore_*.exe := PAIR1 [assign=1,resource=MEMUSE,policy=PSEUDOBALANCED,pri ority=idle]
C:\Documents and Settings\MIKE\Application Data\Folding@home-GPU_card*\Fahcore_*.exe := PAIR0::CPU1+PAIR1::CPU1 [priority=belownormal]
ppd values of wu's......... gpu2 gpu1 gpu0 smp2 smp1 total
511+480+511+1940+1940 4599 5760 4245 1866 1902 18372
480+384+480+1940+1940 6283 5623 5760 1849 1898 21413
384+384+384+1940+1940 6259 5623 5623 1835 1915 21255

---

Way to go, OC!
Man, you've got that XP rig tweaked very nicely now. And I agree entirely with the kudos for Baobab - it's a nice program - and once we get a few configs that kick ass for different OS/CPU combos, we can deploy quick and easy. Thanks for testing and sharing your results.