So not really new to F@H, but as I haven't used F@H in about a decade I figured this is as good of a place as any.
So I installed v7, got some WU, and watched how F@H preformed. And to my surprise my GPU was not at a stable load, jumping all over the place. The average of which is about 50% utilization.
Now I've as of late been using BOINC, and BOINC has been saying for a while now that the CPU-WUs can strangle the little CPU usage needed to support the 10-20x time better preforming GPU-WUs. So I found the running process (FahCore_11.exe) that seemed to be coordinating the GPU work loads, and gave it "AboveNormal" priority. FahCore_11.exe still is using less than <1% of my CPU, but my GPU is now running stable at 93% usage, and my time to complete GPU-WU fell.
So I went to find out how to set this up in the Configuration, Read the FAQ, the Proformanc FAQ..... And could find nothing.
The Only thing I can find is how to slightly elevate both CPU and GPU priority, which isn't the point.
How to I set GPU work loads to have a Slightly higher priority than CPU work loads?
CPU priority for GPU workloads
Moderators: Site Moderators, FAHC Science Team
-
- Posts: 2948
- Joined: Sun Dec 02, 2007 4:36 am
- Hardware configuration: Machine #1:
Intel Q9450; 2x2GB=8GB Ram; Gigabyte GA-X48-DS4 Motherboard; PC Power and Cooling Q750 PS; 2x GTX 460; Windows Server 2008 X64 (SP1).
Machine #2:
Intel Q6600; 2x2GB=4GB Ram; Gigabyte GA-X48-DS4 Motherboard; PC Power and Cooling Q750 PS; 2x GTX 460 video card; Windows 7 X64.
Machine 3:
Dell Dimension 8400, 3.2GHz P4 4x512GB Ram, Video card GTX 460, Windows 7 X32
I am currently folding just on the 5x GTX 460's for aprox. 70K PPD - Location: Salem. OR USA
-
- Posts: 4
- Joined: Sat May 03, 2014 5:35 pm
- Hardware configuration: Intel First Gen i7
Evga GTX280
Occasionally a Second (or third) Evga GTX280, but I loan them to other alot of the time
Re: CPU priority for GPU workloads
P5-133XL wrote:Project lasso works
Pardon My ignorance but that is meaningless in reference to the question. Please enlighten me, as it appears to be some third party log utility.
-
- Posts: 10179
- Joined: Thu Nov 29, 2007 4:30 pm
- Hardware configuration: Intel i7-4770K @ 4.5 GHz, 16 GB DDR3-2133 Corsair Vengence (black/red), EVGA GTX 760 @ 1200 MHz, on an Asus Maximus VI Hero MB (black/red), in a blacked out Antec P280 Tower, with a Xigmatek Night Hawk (black) HSF, Seasonic 760w Platinum (black case, sleeves, wires), 4 SilenX 120mm Case fans with silicon fan gaskets and silicon mounts (all black), a 512GB Samsung SSD (black), and a 2TB Black Western Digital HD (silver/black).
- Location: Arizona
- Contact:
Re: CPU priority for GPU workloads
The current recommendation is to reserve one CPU core to feed the GPU for better performance. The current clients do not have the ability to change priority settings.
Also, core 11 projects were announced as going end of life a while ago. Any gains using a third party priority changer like Lasso will be relatively short lived. A GPU upgrade is in your future.
Also, core 11 projects were announced as going end of life a while ago. Any gains using a third party priority changer like Lasso will be relatively short lived. A GPU upgrade is in your future.
How to provide enough information to get helpful support
Tell me and I forget. Teach me and I remember. Involve me and I learn.
Tell me and I forget. Teach me and I remember. Involve me and I learn.
-
- Posts: 2948
- Joined: Sun Dec 02, 2007 4:36 am
- Hardware configuration: Machine #1:
Intel Q9450; 2x2GB=8GB Ram; Gigabyte GA-X48-DS4 Motherboard; PC Power and Cooling Q750 PS; 2x GTX 460; Windows Server 2008 X64 (SP1).
Machine #2:
Intel Q6600; 2x2GB=4GB Ram; Gigabyte GA-X48-DS4 Motherboard; PC Power and Cooling Q750 PS; 2x GTX 460 video card; Windows 7 X64.
Machine 3:
Dell Dimension 8400, 3.2GHz P4 4x512GB Ram, Video card GTX 460, Windows 7 X32
I am currently folding just on the 5x GTX 460's for aprox. 70K PPD - Location: Salem. OR USA
Re: CPU priority for GPU workloads
Sorry, Process lasso is the correct 3rd party tool I was referring to. It allows you to assign a priority to a core and it will stick between runs.larek wrote:P5-133XL wrote:Project lasso works
Pardon My ignorance but that is meaningless in reference to the question. Please enlighten me, as it appears to be some third party log utility.
-
- Posts: 4
- Joined: Sat May 03, 2014 5:35 pm
- Hardware configuration: Intel First Gen i7
Evga GTX280
Occasionally a Second (or third) Evga GTX280, but I loan them to other alot of the time
Re: CPU priority for GPU workloads
7im wrote:The current recommendation is to reserve one CPU core to feed the GPU for better performance.
Well I could do that by hand too, but its considerably more CPU cycle wasteful. How does one do that in the Client? I know that Threads=CPUcores - 1 does not free up enough CPU to unbind the CPU (it was the first thing I tried)
So how does one set that up using the client?
Gee sorry for donating the use of my not-that-long-ago state-of-the-art, and still under warranty, and still supported by the manufacture, GPUs.7im wrote:Also, core 11 projects were announced as going end of life a while ago. A GPU upgrade is in your future.
Yes I know 7im didn't decide that, and no I not looking for an argument/discussion on obsoletion policies, This just hit me by surprise, I run WU on MUCH older GPU's for other projects, and still think of my cards as Near state-of-the-art, and that they can more than push my 2560x1600 display just fine.
It was since been pointed out that my OS is obsolete (XP) and that core_11 maybe used to to GPU+OS and that a different core_XX is used with my GPU+NewerOS. Is there a link I can get pointed to about GPU's, OS's and Core_XX support?
Re: CPU priority for GPU workloads
FahCore_11 and {_15 /_16} support two types of GPUs: NVidia and ATI(AMD) using some pretty old supporting code. GPU drivers are updated much more frequently than drivers for other hardware both because the hardware upgrade cycle is frequent and because they're always finding things that can be improved. Unfortunately, that means FAH's ability to depend on drivers which go out-of-date frequently provides a variety of challenges not encountered elsewhere.
FAH decided to migrate to OpenCL for two reasons.
1) That's the only API code supported by AMD(ATI)
2) It's supposedly supported by all GPU manufacturers (though the reliability of their support varies widely.)
Recent FAH developments for GPUs have been aimed at the top-of-the-line hardware and the ability to contribute with low-end GPUs has suffered. I know the Pande Group is aware of this trend and that there is an untapped (or at least poorly tapped) resource of low- to medium-end GPUs. I've chatted with the FAH principals about this several times. I'm hopeful something will come of those chats but unfortunately I'm not able to make any promises about when (or even "if") something might happen.
On the other hand, we have to accept a degree of realism. I don't expect any Gamer would be interested in buying any of my HD 5450 GPUs so I do have my doubts about whether Stanford will either. After all, they're approaching 5 years old now, and in dog years, that's past retirement age.
FAH decided to migrate to OpenCL for two reasons.
1) That's the only API code supported by AMD(ATI)
2) It's supposedly supported by all GPU manufacturers (though the reliability of their support varies widely.)
Recent FAH developments for GPUs have been aimed at the top-of-the-line hardware and the ability to contribute with low-end GPUs has suffered. I know the Pande Group is aware of this trend and that there is an untapped (or at least poorly tapped) resource of low- to medium-end GPUs. I've chatted with the FAH principals about this several times. I'm hopeful something will come of those chats but unfortunately I'm not able to make any promises about when (or even "if") something might happen.
On the other hand, we have to accept a degree of realism. I don't expect any Gamer would be interested in buying any of my HD 5450 GPUs so I do have my doubts about whether Stanford will either. After all, they're approaching 5 years old now, and in dog years, that's past retirement age.
Posting FAH's log:
How to provide enough info to get helpful support.
How to provide enough info to get helpful support.
-
- Site Moderator
- Posts: 6986
- Joined: Wed Dec 23, 2009 9:33 am
- Hardware configuration: V7.6.21 -> Multi-purpose 24/7
Windows 10 64-bit
CPU:2/3/4/6 -> Intel i7-6700K
GPU:1 -> Nvidia GTX 1080 Ti
§
Retired:
2x Nvidia GTX 1070
Nvidia GTX 675M
Nvidia GTX 660 Ti
Nvidia GTX 650 SC
Nvidia GTX 260 896 MB SOC
Nvidia 9600GT 1 GB OC
Nvidia 9500M GS
Nvidia 8800GTS 320 MB
Intel Core i7-860
Intel Core i7-3840QM
Intel i3-3240
Intel Core 2 Duo E8200
Intel Core 2 Duo E6550
Intel Core 2 Duo T8300
Intel Pentium E5500
Intel Pentium E5400 - Location: Land Of The Long White Cloud
- Contact:
Re: CPU priority for GPU workloads
Welcome to the F@H Forum larek,
Windows (http://folding.stanford.edu/home/guide/ ... ide/#ntoc4)
Linux (http://folding.stanford.edu/home/guide/ ... ide/#ntoc4)
OSX (http://folding.stanford.edu/home/guide/ ... ide/#ntoc4)
Please note that while Microsoft has ended support for Windows XP, the current F@H Client does run on Windows XP along with some FahCores. However, future releases may or may not support Windows XP. Depending on the OS and the hardware you want to use, here are the requirements along with brief details:larek wrote:...It was since been pointed out that my OS is obsolete (XP) and that core_11 maybe used to to GPU+OS and that a different core_XX is used with my GPU+NewerOS. Is there a link I can get pointed to about GPU's, OS's and Core_XX support?
Windows (http://folding.stanford.edu/home/guide/ ... ide/#ntoc4)
Linux (http://folding.stanford.edu/home/guide/ ... ide/#ntoc4)
OSX (http://folding.stanford.edu/home/guide/ ... ide/#ntoc4)
ETA:
Now ↞ Very Soon ↔ Soon ↔ Soon-ish ↔ Not Soon ↠ End Of Time
Welcome To The F@H Support Forum Ӂ Troubleshooting Bad WUs Ӂ Troubleshooting Server Connectivity Issues
Now ↞ Very Soon ↔ Soon ↔ Soon-ish ↔ Not Soon ↠ End Of Time
Welcome To The F@H Support Forum Ӂ Troubleshooting Bad WUs Ӂ Troubleshooting Server Connectivity Issues