I was running version 6.12.34 on my Mac (OSX 10.5.8), running both Seti@home and World Community Grid. I kept getting the "Waiting for GPU" message on many of the tasks, so I downloaded Boinc version 6.13.12 to see if that might fix the problem.

I no longer seem to be getting the GPU wait, but now I keep getting "Waiting to Run (Scheduler wait)" message on most of my Seti@home tasks. World Community Grid seems fine and runs without an issue.

I've been messing with my settings off and on today and I don't seem to be able to get a combination that will allow my Seti@home task to run for more than a couple of minutes at a time before going into Scheduler Wait. I have one due tomorrow and I'd like it to get done, but it keeps stopping itself.

Any tips on what settings I have wrong? Or is there some problem I'm not aware of?

Thanks,
Cynthia

here's my global_prefs.xml file:

<global_preferences>
<source_project>http://setiathome.berkeley.edu/</source_project>
<source_scheduler>http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi</source_scheduler>

<mod_time>1322507518</mod_time>
<run_on_batteries>1</run_on_batteries>
<run_if_user_active>1</run_if_user_active>
<run_gpu_if_user_active>1</run_gpu_if_user_active>
<idle_time_to_run>3</idle_time_to_run>
<suspend_if_no_recent_input>0</suspend_if_no_recent_input>
<suspend_cpu_usage>0</suspend_cpu_usage>
<leave_apps_in_memory>0</leave_apps_in_memory>
<cpu_scheduling_period_minutes>120</cpu_scheduling_period_minutes>
<max_cpus>16</max_cpus>
<max_ncpus_pct>100</max_ncpus_pct>
<cpu_usage_limit>60</cpu_usage_limit>
<disk_max_used_gb>10</disk_max_used_gb>
<disk_min_free_gb>0.5</disk_min_free_gb>
<disk_max_used_pct>50</disk_max_used_pct>
<disk_interval>60</disk_interval>
<vm_max_used_pct>50</vm_max_used_pct>
<ram_max_used_busy_pct>50</ram_max_used_busy_pct>
<ram_max_used_idle_pct>75</ram_max_used_idle_pct>
<work_buf_min_days>0</work_buf_min_days>
<work_buf_additional_days>0.3</work_buf_additional_days>
<confirm_before_connecting>0</confirm_before_connecting>
<hangup_if_dialed>0</hangup_if_dialed>
<max_bytes_sec_down>0</max_bytes_sec_down>
<max_bytes_sec_up>0</max_bytes_sec_up>
<daily_xfer_limit_mb>0</daily_xfer_limit_mb>
<daily_xfer_period_days>0</daily_xfer_period_days>
<dont_verify_images>0</dont_verify_images>
</global_preferences>

and my global_prefs.override.xml

<global_preferences>
<run_on_batteries>1</run_on_batteries>
<run_if_user_active>1</run_if_user_active>
<run_gpu_if_user_active>1</run_gpu_if_user_active>
<suspend_cpu_usage>0.000000</suspend_cpu_usage>
<start_hour>0.000000</start_hour>
<end_hour>0.000000</end_hour>
<net_start_hour>0.000000</net_start_hour>
<net_end_hour>0.000000</net_end_hour>
<leave_apps_in_memory>1</leave_apps_in_memory>
<confirm_before_connecting>0</confirm_before_connecting>
<hangup_if_dialed>0</hangup_if_dialed>
<dont_verify_images>0</dont_verify_images>
<work_buf_min_days>0.200000</work_buf_min_days>
<work_buf_additional_days>0.300000</work_buf_additional_days>
<max_ncpus_pct>100.000000</max_ncpus_pct>
<cpu_scheduling_period_minutes>14.000000</cpu_scheduling_period_minutes>
<disk_interval>60.000000</disk_interval>
<disk_max_used_gb>10.000000</disk_max_used_gb>
<disk_max_used_pct>80.000000</disk_max_used_pct>
<disk_min_free_gb>0.500000</disk_min_free_gb>
<vm_max_used_pct>50.000000</vm_max_used_pct>
<ram_max_used_busy_pct>50.000000</ram_max_used_busy_pct>
<ram_max_used_idle_pct>90.000000</ram_max_used_idle_pct>
<max_bytes_sec_up>0.000000</max_bytes_sec_up>
<max_bytes_sec_down>0.000000</max_bytes_sec_down>
<cpu_usage_limit>60.000000</cpu_usage_limit>
<daily_xfer_limit_mb>0.000000</daily_xfer_limit_mb>
<daily_xfer_period_days>0</daily_xfer_period_days>
</global_preferences>

The "scheduler wait" message in 6.13 is essentially the same as the "waiting for memory" message in 6.12; a GPU does not have enough memory to continue the work.

But in 6.12 it would also come up when other causes left no memory to be used. Essentially, what this message now means is that the application was temporarily exited by BOINC and is waiting to be rescheduled, to run again at a later time when we hope enough memory is available.

Which begs the questions:
1. Do you have a GPU? If so, what kind?
2. How much memory does the GPU have?
3. How much memory does your Mac have?
4. How many CPUs does your Mac have?

---

Hmm okay, looking... I have an Nvidia GeForce 9400M. I'm not sure if that's GPU or not, googling tells me that a 9400M G is a GPU. The Type just says Display on my Mac, doesn't say GPU like in many of the examples. So perhaps I don't have GPU.

If not, I'll have to find where to set Boinc so that I don't get units that require GPU processing... or so I assume.

I assume VRam is memory on the card, it says 256MB

My Mac has one processor, 2 cores, 2 GB memory.

It's just the baseline MacBook without anything fancy. It's about 4 years old now.
Thanks for your help.
Cynthia

I have an Nvidia GeForce 9400M.

That's a GPU that you can use to do work with on some projects. In a sense everyone has a GPU, as that's the graphical processing unit. Without it you don't see anything on the screen. However, since a couple of years one is able to use the ATI/AMD and nVidia GPUs for calculations, just as you would use a CPU.

I assume VRam is memory on the card, it says 256MB

Yes, VRAM is Video RAM, memory on the videocard. 256MB is not enough to run work on the majority of projects. Seti needs at least 384MB on the videocard to be able to run a task to completion, especially when you use the latest drivers.

You may have to look around and ask on the various projects if they support a card with this little memory. If not, best thing is to disable the use of it on those projects that don't. You can normally do so in the project preferences, in the case of Seti that's at http://setiathome.berkeley.edu/prefs.php?subset=project; go there and set "Use Nvidia GPU?" to No.

You can do so on other projects as well. Using the GPU or not is a project preference, not a global one as not all projects support GPU use yet.

---

Ah okay, thanks so much for your help. So many preference screens. :) I changed my Seti&Home preferences to not get any GPU units.

Cynthia

Seti is down for their scheduled Tuesday maintenance period. Should be back up in an hour or two, but will be very slow for awhile.

Cynthia, I had the same "Waiting to Run (Scheduler wait)" issue with Einstein@home on my late-2008 MacBook Pro (it seems like that project needs more than 300 MB of free GPU memory). I then tried the PrimeGrid project, and it seems to work fine on both GPUs, Nvidia 9400M (256MB) and Nvidia 9600M (512MB). So you could try PrimeGrid if you want BOINC to utilize your GPU.

(BTW, I didn't know SETI@home had GPU support for Mac, BOINC wiki says it doesn't: http://boinc.berkeley.edu/wiki/GPU_computing Maybe they should update that wiki).

(BTW, I didn't know SETI@home had GPU support for Mac, BOINC wiki says it doesn't: http://boinc.berkeley.edu/wiki/GPU_computing Maybe they should update that wiki).

Seti doesn't have a Stock Cuda app for OSX, and Lunatics hasn't even done a Cuda port for OSX, so the question is why would Boinc Suspend Seti CPU tasks because there wasn't enough GPU memory?

Claggy

(BTW, I didn't know SETI@home had GPU support for Mac, BOINC wiki says it doesn't: http://boinc.berkeley.edu/wiki/GPU_computing Maybe they should update that wiki).

Seti doesn't have GPU applications by default for the Mac or Linux. But you can download a 3rd party app and used that through the anonymous platform, just as you can use a plethora of optimized apps through the anonymous platform. Doesn't mean that the project then has such an app, as when you add the project to BOINC it won't automatically download the app.

---