HCC GPU Launched.

[Snow Crash]

Repost from page 11. This config will run 2 GPUs and also run HCC on as many CPU cores as you have left over. The indicator if you can/ should run multiple GPU instances is if your GPU utilization is less than roughly 95%. After that you will still get some increase in work done but the machine will respond slowly (laggy).

<app_info>
   <app>
      <name>hcc1</name>
      <user_friendly_name>Help Conquer Cancer</user_friendly_name>
   </app>

   <file_info>
      <name>wcg_hcc1_img_6.56_windows_intelx86__nvidia_hcc1</name>
      <executable/> 
   </file_info>
   <file_info>
      <name>hcckernel.cl.6.56</name>
      <executable/>
   </file_info>

   <file_info>
      <name>wcg_hcc1_img_6.56_windows_intelx86</name>
      <executable/>
   </file_info>


   <app_version>
      <app_name>hcc1</app_name>
      <version_num>656</version_num>
      <platform>windows_intelx86</platform>
      <plan_class>nvidia_hcc1</plan_class>
      <avg_ncpus>1.0</avg_ncpus>
      <max_ncpus>0.783972</max_ncpus>
      <coproc>
         <type>CUDA</type>
         <count>.5</count>
      </coproc>
      <file_ref>
          <file_name>wcg_hcc1_img_6.56_windows_intelx86__nvidia_hcc1</file_name>
          <main_program/>
      </file_ref>
      <file_ref>
         <file_name>hcckernel.cl.6.56</file_name>
         <open_name>hcckernel.cl</open_name>
      </file_ref>
   </app_version>

<app_version>
    <app_name>hcc1</app_name>
    <version_num>656</version_num>
    <platform>windows_intelx86</platform>
    <avg_ncpus>1.000000</avg_ncpus>
    <max_ncpus>1.000000</max_ncpus>
    <flops>3231932486.733310</flops>
    <api_version>6.13.0</api_version>
    <file_ref>
        <file_name>wcg_hcc1_img_6.56_windows_intelx86</file_name>
        <main_program/>
    </file_ref>
</app_version>

</app_info>

[bluestang]

Ok, thanks guys for all the info. I finally have 3 GPU HCC WUs running and 5 CPU GFAM WUs running.

Should be a productive night

Here is my app_info for it:

<app_info>
   <app>
      <name>hcc1</name>
      <user_friendly_name>Help Conquer Cancer</user_friendly_name>
   </app>
   <file_info>
      <name>wcg_hcc1_img_6.56_windows_intelx86__ati_hcc1</name>
      <executable/> 
   </file_info>
   <file_info>
      <name>hcckernel.cl.6.56</name>
      <executable/>
   </file_info>
      <app_version>
      <app_name>hcc1</app_name>
      <version_num>656</version_num>
      <platform>windows_intelx86</platform>
      <plan_class>ati_hcc1</plan_class>
      <avg_ncpus>1.0</avg_ncpus>
      <max_ncpus>1.0</max_ncpus>
      <coproc>
         <type>ATI</type>
         <count>.33</count>
      </coproc>
      <file_ref>
          <file_name>wcg_hcc1_img_6.56_windows_intelx86__ati_hcc1</file_name>
          <main_program/>
      </file_ref>
      <file_ref>
         <file_name>hcckernel.cl.6.56</file_name>
         <open_name>hcckernel.cl</open_name>
      </file_ref>
   </app_version>

<app>
    <name>gfam</name>
    <user_friendly_name>GO Fight Against Malaria</user_friendly_name>
</app>

   <file_info>
      <name>wcgrid_gfam_vina_6.12_windows_x86_64</name>
      <executable/>
   </file_info> 
   <file_info>
      <name>wcgrid_gfam_vina_prod_x86_64.exe.6.12</name>
      <executable/>
   </file_info>
   <file_info>
      <name>wcgrid_gfam_gfx_prod_x86_64.exe.6.12</name>
       <executable/>
   </file_info>

<app_version>
    <app_name>gfam</app_name>
    <version_num>612</version_num>
    <platform>windows_x86_64</platform>
    <avg_ncpus>1.0</avg_ncpus>
    <max_ncpus>1.0</max_ncpus>
    <flops>3347548492.458962</flops>
    <api_version>7.1.0</api_version>
    <file_ref>
        <file_name>wcgrid_gfam_vina_6.12_windows_x86_64</file_name>
        <main_program/>
    </file_ref>
    <file_ref>
         <file_name>wcgrid_gfam_vina_prod_x86_64.exe.6.12</file_name>
          <open_name>AutoDockVina64.exe</open_name>
     </file_ref>
     <file_ref>
         <file_name>wcgrid_gfam_gfx_prod_x86_64.exe.6.12</file_name>
          <open_name>graphics_app</open_name>
     </file_ref>
</app_version>

<app_version>
    <app_name>gfam</app_name>
    <version_num>612</version_num>
    <platform>windows_intelx86</platform>
    <avg_ncpus>1.000000</avg_ncpus>
    <max_ncpus>1.000000</max_ncpus>
    <flops>3347548492.458962</flops>
    <api_version>7.1.0</api_version>
    <file_ref>
        <file_name>wcgrid_gfam_vina_6.12_windows_x86_64</file_name>
        <main_program/>
    </file_ref>
    <file_ref>
         <file_name>wcgrid_gfam_vina_prod_x86_64.exe.6.12</file_name>
          <open_name>AutoDockVina64.exe</open_name>
     </file_ref>
     <file_ref>
         <file_name>wcgrid_gfam_gfx_prod_x86_64.exe.6.12</file_name>
          <open_name>graphics_app</open_name>
     </file_ref>
</app_version>
</app_info>

[Mydog]

the key is you do one GPU WU if you have a "mild" vid card like yours and let the cpu's do their work BUT if you have a "BIG" vid card you use that script and multiple GPU WU and let it fly..
Mydog broke 500,000 WCG points today on just GPU on his 3770K system

True and it's only one GPU in that system a 7970 but I hope to get two more this weekend. Might need to change the PSU as I'm now running on a 450W cheap PSU, not sure it'll handle another two 7970's

Nice "little" increase in output going from CPU/GPU WU's to GPU only.



Edit:
Just bumped the clocks from 925 to 1000 on core and got the total time below 2 mins on the WU's, average looks like 116 sec from the 28 first WU's.

[fallwind]

A 7870 came online tonight in the wifes rig. 2x 7970's (hopefully) coming online tomorrow! I just about soiled my shorts when I saw how fast the 7870 chewed through the WU's compared to my GTX 460's....lol! I'm feeding it with 6 cores of a 980x at 3.8GHz but the GPU still goes idle once in a while. I'll bump it to 8 cores if it runs stable and error free overnight.

[Mydog]

A 7870 came online tonight in the wifes rig. 2x 7970's (hopefully) coming online tomorrow! I just about soiled my shorts when I saw how fast the 7870 chewed through the WU's compared to my GTX 460's....lol! I'm feeding it with 6 cores of a 980x at 3.8GHz but the GPU still goes idle once in a while. I'll bump it to 8 cores if it runs stable and error free overnight.

you're running 6 WU's simultaneously on that 7870?? What times are you getting on those WU's?

[Rob_B]

I'm still having trouble running more than one WU at the time on my 7770. Last night I tried again and it looked good for some time. But when I got home a little later it got stuck on 2 WU's again. Lost 2.5 hours on that rig.

Any idea what causes this. I'm on the latest ATI drivers.

[fallwind]

@ Mydog,

Just taking a small sample of elapsed times shown in BOINC, it looks like 3:30 to 3:50. So roughly 30-38 seconds per WU.

[Mydog]

@ Mydog,

Just taking a small sample of elapsed times shown in BOINC, it looks like 3:30 to 3:50. So roughly 30-38 seconds per WU.

Looks like I need to test with more than 4 WU's simultaneously on my 7970 then, WU's now use just short of 120 sec on average total which is 30 seconds per WU.

I assume that for 6 WU's you set the number .166 right?

[Aerou]

GJ guys!

Something tells me this would be the perfect time to have all the APP_INFO tags for each project in one place
Can anyone please write post the tags for FAAH?

[Mydog]

I'm now testing to run 6 WU's simultaneously on my 7970 and I found something interesting. Total time per WU went up as expected going from 4 to 6 WU's and the increase per hour where slim, from 120 WU's an hour to 130+, but from what I can see in the WCG page points claimed and granted went up from an average of 30 point per WU to 35(this is Boinc points). CPU usage went from 50% max(4 GPU WU's) to 75% max(6 GPU WU's)
Can anyone else confirm these findings?

[Fulen]

Testing the app_info file on 7770 and it seems to claim more (Q9650 3.6MHz 2 7770 running 4 GPU tasks)

Claimed/ Granted
GPU only
51.3 / 36.8
42.0 / 37.6
39.0 / 34.4
41.8 / 37.2
48.9 / 33.7
49.2 / 34.1
46.3 / 31.9
CPU + GPU
33.3 / 34.5
36.4 / 36.4
32.5 / 32.2
32.3 / 32.4
36.4 / 35.6
32.2 / 31.9

[bluestang]

I was running 3 GPU WUs on my 7870 last night for a couple hours and no issues. Went to bed and woke up to a BSOD. Not sure how long it was like that

I noticed 2 of the WUs had "Computational Error" status when I got back into BOINC. Not sure what happened, didn't have much time other than to re-boot and change back to 2 GPU WUs before I left for work.

I'm gonna try and go back to 3 tonight when I get home from work

I did notice I had a hard time keeping my cache full of HCC GPU WUs thought when running 3 at a time...is there a trick to keeping it full (.01 min/4.0 max buffer set) other than manually hitting the update button when you can??? Sorta hard to do that when you sleeping or at work.

EDIT: I have a 7770 coming for my Work PC (old HD 3650 died) and 2x7770 for my Home PC. Not sure if I'm sending the 7870 back when I get the 2x7770 or keeping it. More cards than slots right now

[Mydog]

Testing the app_info file on 7770 and it seems to claim more (Q9650 3.6MHz 2 7770 running 4 GPU tasks)

Claimed/ Granted
GPU only
51.3 / 36.8
42.0 / 37.6
39.0 / 34.4
41.8 / 37.2
48.9 / 33.7
49.2 / 34.1
46.3 / 31.9
CPU + GPU
33.3 / 34.5
36.4 / 36.4
32.5 / 32.2
32.3 / 32.4
36.4 / 35.6
32.2 / 31.9

From this I can read that WU's takes about 3 mins+- total right? two on each of the two 7770's that gives you 80 WU per hour more or less?

[=[PULSAR]=]

Is anyone else have trouble downloading new WU's? I woke up this morning to nothing running and when I update the project its not able to attain any new ones.

[Mydog]

Is anyone else have trouble downloading new WU's? I woke up this morning to nothing running and when I update the project its not able to attain any new ones.

You need to set the profile in WCG to run GPU I only or it might depend on the app_info file you're using.

[=[PULSAR]=]

It was running great when I went to bed. I removed the app_info and started grabbing WU, but after replacing the app_info still a no go.

[Mydog]

Are you running GPU only or GPU/CPU?

Guys when you talk about load on your GPU's are you talking about the load when it's working on a WU or a percentage of time the GPU has work? In my case on the 7970 looks like this 94% load 85% of the time running 6 WU's simultaneously and CPU(3770K @4,7GHz) at 77% max load.

Might have room for two more WU's on the GPU.

WU's seem to vary a lot in weight(Time it takes vs points claimed/granted)

[=[PULSAR]=]

GPU only, but got it working now. Unattached the project and added it again seems to be working now.

[ggdh]

So what sort of numbers should I expect from GTX580? Damn expensive to run it 24/7 but I may be in for SCW...

[Fulen]

From this I can read that WU's takes about 3 mins+- total right? two on each of the two 7770's that gives you 80 WU per hour more or less?

Av. ~3,15
Next up is overclocking GPU`s

[=[PULSAR]=]

Nevermind problem isn't fixed. I'll reattach project and it will give me 17 WU but it will just run out and never fetch more

Here is the app_info I'm using

<app_info> 
<app>
    <name>hcc1</name>
    <user_friendly_name>Help Conquer Cancer</user_friendly_name>
</app>
<file_info>
    <name>wcg_hcc1_img_6.56_windows_intelx86__nvidia_hcc1</name>
    <nbytes>3499008.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <executable/>
    <signature_required/>
    <file_signature>
792397d3ec34c91119da45507eb1f8e732452b9d40c1d4adc313cb4d0b6db671
0382de7f034a2dea10b75e2ab2e012b1ce9c3effa738a48f50daa615fa3e6b82
3f4e28d9986968d347552f0298571f3755a601f8834ab9f554d457ea2f0a71c4
843ef8c8b5e4b04633591321627f9a403f12f380bcb5702593f2e42ca0ea5514
.
</file_signature>
</file_info>
<file_info>
    <name>hcckernel.cl.6.56</name>
    <nbytes>63144.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
a2633e96b0df14406470c5c98e49533b903776e06356cf2dfad1ee853eb15c42
190a29639bed90e5a02c426669ff46441f90f1dad3fe303f05314befeffc28e5
7433da8e2545665f9900ec714a6f9f2d83246de5c32c6c060ac017d1392e0ed6
34839df4b734c410bb5007fe6108d793be40665a4c538654240dc28055bc01fe
.
</file_signature>
</file_info>
<file_info>
    <name>hcc1_image01_6.56.tga</name>
    <nbytes>262188.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
7117813faaff1bceb99d172432ff67417397b809172fc09c47a0db414106c68d
de9bf88ed64f518e78b5e7ab79637335c9b166003bd3f32a29915101c1758746
d1db7a90bc201c35b1c65fcfb2a9b091253b1979bc6f9db748feb5a6e9d5c804
cd875399b2ee4016fc210f5b76f15aa0902477ce1c0a945fb56c178a0d65de4c
.
</file_signature>
</file_info>
<file_info>
    <name>hcc1_image01_6.56.tga</name>
    <nbytes>262188.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
7117813faaff1bceb99d172432ff67417397b809172fc09c47a0db414106c68d
de9bf88ed64f518e78b5e7ab79637335c9b166003bd3f32a29915101c1758746
d1db7a90bc201c35b1c65fcfb2a9b091253b1979bc6f9db748feb5a6e9d5c804
cd875399b2ee4016fc210f5b76f15aa0902477ce1c0a945fb56c178a0d65de4c
.
</file_signature>
</file_info>
<file_info>
    <name>hcc1_image02_6.56.tga</name>
    <nbytes>32812.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
0d3314db1591dcf067aa6551e3582f5274ff10287fb384c11934a98245fd1631
be5249db0ae6d0138b191618ae9502e853a681f879e7fa3c94fec8effdd6685b
a325ec7538ec8abb1fd719629c691bf721da031a87cf69d20b0f501a38e16a92
dbdd7442d90b64f30deff58bc46c5096fb017f4d6410eb8e9bfacb09b2c6a8ae
.
</file_signature>
</file_info>
<file_info>
    <name>hcc1_image03_6.56.tga</name>
    <nbytes>131116.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
58915b00c9fae53d18ef72f179c6ff1222c134b0c0e696c3bbfddb91c650b26f
c5856311eea2fe2128d9dbdd558b47963332cf9f9ae910689a133b03e7113715
39babe2ce7303b73e9bfc69b1f01cc4199c7dfab13322d5b40adcb8ee167832d
81df23835ba4e12430a50aeb534b8bfed9a82ce54fb0a4018104f1adb36778d9
.
</file_signature>
</file_info>
<file_info>
    <name>hcc1_image04_6.56.tga</name>
    <nbytes>262188.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
59882941f8d7b844b3ac20301fba70c98ab9cedd5a41855d24b0721b6b6275e6
f1de708d8b43ae7f0198f21924f528c1c710b6d36776d881fee6ebb539cf2f4e
4c4f63997be3eb6157b9a3a5922bdbdebb26fc1cfeae4d13980682800c697ffa
54a7a64cc14694ec457132320f0c2bcc9c3fed675f71cbe388d9ec55332c1bae
.
</file_signature>
</file_info>
<app_version>
    <app_name>hcc1</app_name>
    <version_num>656</version_num>
    <platform>windows_intelx86</platform>
    <avg_ncpus>1.000000</avg_ncpus>
    <max_ncpus>0.783972</max_ncpus>
    <flops>56530729222.498611</flops>
    <plan_class>nvidia_hcc1</plan_class>
    <api_version>7.1.0</api_version>
    <file_ref>
        <file_name>wcg_hcc1_img_6.56_windows_intelx86__nvidia_hcc1</file_name>
        <main_program/>
    </file_ref>
    <file_ref>
        <file_name>hcckernel.cl.6.56</file_name>
        <open_name>hcckernel.cl</open_name>
    </file_ref>
    <file_ref>
        <file_name>hcc1_image01_6.56.tga</file_name>
        <open_name>boinc_wcg_skin_w-logo.tga</open_name>
    </file_ref>
    <file_ref>
        <file_name>hcc1_image02_6.56.tga</file_name>
        <open_name>pbibmablk.tga</open_name>
    </file_ref>
    <file_ref>
        <file_name>hcc1_image03_6.56.tga</file_name>
        <open_name>UHN_stacked2.tga</open_name>
    </file_ref>
    <file_ref>
        <file_name>hcc1_image04_6.56.tga</file_name>
        <open_name>HCC_LOGO.tga</open_name>
    </file_ref>
    <coproc>
        <type>CUDA</type>
        <count>0.5000000</count>
    </coproc>
    <gpu_ram>266338304.000000</gpu_ram>
</app_version>
</app_info>

[Mydog]

@Pulsar
Where did you get that one?

I'm using this one for ATI GPU, running on GPU only

<app_info>
 <app>
 <name>hcc1</name>
<user_friendly_name>Help Conquer Cancer</user_friendly_name>
 </app>
 <file_info>
 <name>wcg_hcc1_img_6.56_windows_intelx86__ati_hcc1</name>
 <executable/>
 </file_info>
 <file_info>
 <name>hcckernel.cl.6.56</name>
 <executable/>
 </file_info>
 <app_version>
 <app_name>hcc1</app_name>
 <version_num>656</version_num>
 <platform>windows_intelx86</platform>
<plan_class>ati_hcc1</plan_class>
 <avg_ncpus>1.0</avg_ncpus>
 <max_ncpus>1.0</max_ncpus>
 <coproc>
 <type>ATI</type>
 <count>[COLOR="#FF0000"].5[/COLOR]</count>
 </coproc>
 <file_ref>
 <file_name>wcg_hcc1_img_6.56_windows_intelx86__ati_hcc1</file_name>
 <main_program/>
 </file_ref>
 <file_ref>
 <file_name>hcckernel.cl.6.56</file_name>
 <open_name>hcckernel.cl</open_name>
 </file_ref>
 </app_version>
</app_info> 

The red number indicates how many WU's to run like this 1=1, .5=2, .33=3 and .25=4 etc.

+ you need to set up the profile for that rig like this in your WCG page:

[OldChap]

Just so I understand this correctly.... Running high end card with 4 instances uses 4 cores cpu and 2 cards would use 8 cpu cores right?

So with 7 slots, 7 x 7970's and some flexi risers Dave could run 28 cores of his 32 core beast on this??? I wonder what that would produce from one $8000 rig?

[Mydog]

I'm not sure, not tested yet, but I think you can run up to 8 WU's simultaneously on one 7970 which will use all cores and threads on a 3770K. I'm running 6 WU's on a single 7970 ATM, been so all day, and it seems to be working perfect so far.

[Snow Crash]

^^ are you still gaining efficiencies from running more than 4 instances at once?