Maybe lack of CPU core availability? The main type of GPU work currently available (which is new) is distributed with a stated need for a full CPU core for support (unlike, say, 0.2). Depending on other settings and on competing needs for work, this could be the difference.
Also the current application requires a lot of GPU RAM. I don't know whether that leads to multiplicity reduction by boinc in some cases. As you choose to hide your computers, I can't give a more informed comment.
Same thing happened to me shauge , only 1 gpu at a time from 3. I updated drivers (376.33) on all computers , updated BOINC Manager to latest (7.6.33)....and walla I started running 3 at a time again.
I am sorry to disturb. I have the same problem. I used to run with factor 0.33 so that there were 3 gpu tasks at a time running. Now with new apps one of my hosts does only 1 gpu task at a time. I checked the gpu utilisation factor for all 3 gpu apps. It is and was set to 0.33.
I also have enough cores to run 3 pgu task like (1 cpu 0,33 nvidia gpu). I have 8 cores.
Still my host does like this: 8 Multi-Directed Continuous Gravitational Wave search CV
Check if your BOINC client running E@H in "high priority mode" (on tasks tab). It may be cause if you right about 8 (not 7) Multi-Directed Continuous Gravitational Wave search CV tasks running on 8 cores CPU.
If so - try to reduce work cache size in BOINC settings and(only after new cache size, 1 day max) abort some of GW WUs until BOINC stop running in ""high priority". Or just wait few days while it clear some of WU queue without downloading new ones.
If it not help - we can always bring "the big guns" and use app_info config file to manuallydistribute tasksand priorities.
The fix for me was to change the gpu utilization factor for all 3 gpu apps. However, it did not start doing multiple apps immediately, even after forcing a config update. It may be because I had too much work for the other CPU only project I run. The new gpu apps require a whole CPU core pr. instance. After a day or two it started running multiple GPU apps.
Maybe lack of CPU core
)
Maybe lack of CPU core availability? The main type of GPU work currently available (which is new) is distributed with a stated need for a full CPU core for support (unlike, say, 0.2). Depending on other settings and on competing needs for work, this could be the difference.
Also the current application requires a lot of GPU RAM. I don't know whether that leads to multiplicity reduction by boinc in some cases. As you choose to hide your computers, I can't give a more informed comment.
Go to the project settings
)
Go to the project settings page for your account here. BRP, FGRP, and GW all have separate CPU utilization settings.
If you're using app_config
)
If you're using app_config file you'll have to update the application name in the file.
mmonnin wrote:If you're using
)
Yes, but an app not specified in an app_config.xml will/should use the utilization settings from the website.
Same thing happened to me
)
Same thing happened to me shauge , only 1 gpu at a time from 3. I updated drivers (376.33) on all computers , updated BOINC Manager to latest (7.6.33)....and walla I started running 3 at a time again.
Ok, thanks for the answers. I
)
Ok, thanks for the answers. I had missed that there have been added GPU utilization for each of the various GPU apps.
I am sorry to disturb. I have
)
I am sorry to disturb. I have the same problem. I used to run with factor 0.33 so that there were 3 gpu tasks at a time running. Now with new apps one of my hosts does only 1 gpu task at a time. I checked the gpu utilisation factor for all 3 gpu apps. It is and was set to 0.33.
I also have enough cores to run 3 pgu task like (1 cpu 0,33 nvidia gpu). I have 8 cores.
Still my host does like this: 8 Multi-Directed Continuous Gravitational Wave search CV
and 1 Gamma-ray pulsar binary search #1 on GPUs
What is wrong?
https://einsteinathome.org/de/host/12462600
Check if your BOINC client
)
Check if your BOINC client running E@H in "high priority mode" (on tasks tab). It may be cause if you right about 8 (not 7) Multi-Directed Continuous Gravitational Wave search CV tasks running on 8 cores CPU.
If so - try to reduce work cache size in BOINC settings and(only after new cache size, 1 day max) abort some of GW WUs until BOINC stop running in ""high priority". Or just wait few days while it clear some of WU queue without downloading new ones.
If it not help - we can always bring "the big guns" and use app_info config file to manually distribute tasks and priorities.
Mad_Max wrote:If it not help
)
app_config.xml file can do that.
The fix for me was to change
)
The fix for me was to change the gpu utilization factor for all 3 gpu apps. However, it did not start doing multiple apps immediately, even after forcing a config update. It may be because I had too much work for the other CPU only project I run. The new gpu apps require a whole CPU core pr. instance. After a day or two it started running multiple GPU apps.