Can someone here help a noob and post an app_info file that will let me run 2 at a time on a GTX 460? I'm only @ 68% GPU, 55c temp and 30% memory running 1.
Thanks in advance.
We implemented this setting precisely to avoid the necessity for an app_info.xml.
App_info is no longer needed. Why are people asking/posting them? If required for other reasons, it would be better not to put them in this sticky news thread.
Can someone here help a noob and post an app_info file that will let me run 2 at a time on a GTX 460? I'm only @ 68% GPU, 55c temp and 30% memory running 1.
Thanks in advance.
We implemented this setting precisely to avoid the necessity for an app_info.xml.
App_info is no longer needed. Why are people asking/posting them? If required for other reasons, it would be better not to put them in this sticky news thread.
I am much impressed by this feature, too. On Linux, I interpret the output of nvidia-smi correctly, then GPU utilisation went up from 55% to 90%, memory doubled as expected.
i have update client but now i got: Message form server: (Reached daily quota of 2 tasks).
why only 2? it was 32 tasks per cpu per day.
that's b/c 4 out of your 5 hosts are erring out on a massive number of WUs. and the one host that doesn't have any errors has a boatload of invalids. at any rate, if a host returns enough erred tasks to the project server in a given time span, a limitation on that host's daily task quota is implemented server-side to prevent the host from causing another massive influx of errors. once you've figured out why you're getting errors and fixed it, you'll begin to return valid results again. the more consecutive valid results you return, the sooner your daily quota will go back up to its normal value.
Hallo!
Since I´m using the new applications, I´ve also lots of crunching errors, much, much more than ever before, and I don´t know really why. They all end up after a wide spread of crunching times with this error code: (0x2) - exit code 2 (0x2). There is no output file.
Futhermore I can crunch only 3 of this S6LV1 files in parallel, whereas it had been before 4 tasks in parallel - i3 2100 dual core with HT -. My adjustments allow for 100% CPU usage.
So I thought it might be a good idea to test the RAM thoroughly, more than by the BIOS during startup. Do you know a tool for doing this ???
If you want BOINC to reserve less, write the factor in there. E.g. if you want to run two tasks on the same GPU, use a factor of 0.5, BOINC will then reserve half a GPU for a task.
it can't be used in case user have several host with different gpu.
for example, some of my hosts can crunch only 2 gpu's tasks at a time, while other can crunch 5 tasks. so i have to mess with appp_info file.
am i right?
If you want BOINC to reserve less, write the factor in there. E.g. if you want to run two tasks on the same GPU, use a factor of 0.5, BOINC will then reserve half a GPU for a task.
it can't be used in case user have several host with different gpu.
for example, some of my hosts can crunch only 2 gpu's tasks at a time, while other can crunch 5 tasks. so i have to mess with appp_info file.
am i right?
yes, the web setting will cause all Einstein@Home hosts w/ GPUs to use the same factor. you will have to continue using a slightly different app_info.xml on each machine if the GPUs in those machines aren't all going to crunch the same number of tasks simultaneously.
Bump RE: RE: Can
)
Bump
App_info is no longer needed. Why are people asking/posting them? If required for other reasons, it would be better not to put them in this sticky news thread.
RE: Bump RE: RE: Can
)
Oh, that is what I have missed, thx for that!
RE: I am much impressed by
)
I tested this and benchmarked the performance. Here are the results
HOW TO - Full installation Ubuntu 11.10
i have update client but now
)
i have update client but now i got: Message form server: (Reached daily quota of 2 tasks).
why only 2? it was 32 tasks per cpu per day.
RE: i have update client
)
that's b/c 4 out of your 5 hosts are erring out on a massive number of WUs. and the one host that doesn't have any errors has a boatload of invalids. at any rate, if a host returns enough erred tasks to the project server in a given time span, a limitation on that host's daily task quota is implemented server-side to prevent the host from causing another massive influx of errors. once you've figured out why you're getting errors and fixed it, you'll begin to return valid results again. the more consecutive valid results you return, the sooner your daily quota will go back up to its normal value.
Hallo! Since I´m using the
)
Hallo!
Since I´m using the new applications, I´ve also lots of crunching errors, much, much more than ever before, and I don´t know really why. They all end up after a wide spread of crunching times with this error code: (0x2) - exit code 2 (0x2). There is no output file.
Futhermore I can crunch only 3 of this S6LV1 files in parallel, whereas it had been before 4 tasks in parallel - i3 2100 dual core with HT -. My adjustments allow for 100% CPU usage.
So I thought it might be a good idea to test the RAM thoroughly, more than by the BIOS during startup. Do you know a tool for doing this ???
Kind regards
Martin
RE: Hallo! So I thought it
)
memtest
RE: On Linux, I interpret
)
Steffen, what card model and what linux do you use?
my nvidia-smi return me much less info:
[pre]
+------------------------------------------------------+
| NVIDIA-SMI 3.295.20 Driver Version: 295.20 |
|-------------------------------+----------------------+----------------------+
| Nb. Name | Bus Id Disp. | Volatile ECC SB / DB |
| Fan Temp Power Usage /Cap | Memory Usage | GPU Util. Compute M. |
|===============================+======================+======================|
| 0. GeForce GTX 560 Ti | 0000:01:00.0 N/A | N/A N/A |
| 72% 61 C N/A N/A / N/A | 81% 825MB / 1023MB | N/A Default |
|-------------------------------+----------------------+----------------------|
| 1. GeForce GTX 560 Ti | 0000:08:00.0 N/A | N/A N/A |
| 30% 29 C N/A N/A / N/A | 0% 5MB / 2047MB | N/A Default |
|-------------------------------+----------------------+----------------------|
| Compute processes: GPU Memory |
| GPU PID Process name Usage |
|=============================================================================|
| 0. Not Supported |
| 1. Not Supported |
+-----------------------------------------------------------------------------+
[/pre]
RE: If you want BOINC to
)
it can't be used in case user have several host with different gpu.
for example, some of my hosts can crunch only 2 gpu's tasks at a time, while other can crunch 5 tasks. so i have to mess with appp_info file.
am i right?
RE: RE: If you want BOINC
)
yes, the web setting will cause all Einstein@Home hosts w/ GPUs to use the same factor. you will have to continue using a slightly different app_info.xml on each machine if the GPUs in those machines aren't all going to crunch the same number of tasks simultaneously.