Hi people, I have recently started using BOINC but was getting a bit frustrated with BOINC manager. I used it in the advanced view.

My problem is I finished one task and it didn't automatically move onto the next one. I would like this to happen so that I don't have to manually tell it to begin another task.

I only want to do one task at a time, because I use my computer for other stuff during it.

Let me know if you can help with this.

Thanks

I just thought I would add to my post above. It wasn't just a single issue I got frustrated with, I generally felt that BOINC manager had been set up in a rushed manner, whilst the main activity of the software is very good. It is not a new piece of software, from what I can see, so seeing bugs and inadequacies did not fill me with confidence that it has a committed team behind it. I also felt not having a proper help service was a problem too. I emailed the head of the project and he thankfully replied, helped me a bit and referred me to here.

I nearly uninstalled the program, but thought I would give it another try.

Best regards and thanks for any help.

You need to give more info about the setup that you are running.

http://boinc.berkeley.edu/dev/forum_thread.php?id=8600

Which BOINC version, which operating system?

What kind of computation settings have you applied via the BOINC Manager?

Hi Richie,

Boinc 7.7.42 (x64)

Wxwidgets version 3.0.1

Windows 7 Home Premium

64 bit operating system

In terms of preferences I have the following:

Run always (well suspended at the moment, but when I want to use it I put it on always)

Processor usage:

Switch between applications every 60 minutes (this was the default setting)
On multi processor systems use at most 0% of the processors
Use at most 100% of CPU time

Network usage:

Skip image file verification: unticked
Confirm before connecting to internet: unticked
Disconnect when done: unticked

Disk and Memory usage:

Leave applications in memory whilst suspended (ticked)

That is all really. Thanks for getting back to me.

Processor usage:

On multi processor systems use at most 0% of the processors

What processor does your computer have? If it is for example a quad core, then you could set that value to 25% and then it should run only one core and only one task for it (normal situation, without any special settings).

If you have a HT (hyper threading) capable CPU and you have enabled HT in bios and use all cores, then the amount of "cores" is same as "threads". "4 core, 8 threads" i7 CPU could use 8 "cores" in BOINC, for example. That way, it would be able to run 8 CPU tasks (if the application requires exactly 1 core per tasks).

Dual core... 50%
Quad core... 25%
4 cores / 8 threads CPU... HT "ON", all cores.... and 1 "core" would be 12%.
6 cores / 12 threads CPU... HT "ON", all cores.... and 1 "core" would be 8%.

Btw. What project and which application from it do you want to run?

Boinc 7.7.42 (x64)

Just for clarification for those helping, is it 7.2.42 or 7.4.42? 7.7.42 does not exist.

---

Sorry people, Ageless is right, I got my product version wrong. I have BOINC 7.4.42.

Will get the rest of the info for you now.

Thanks

Just told Ageless that I got my version of Boinc wrong (duh!) I have version 7.4.42.

My processor is the following:


Processor Intel(R) Core(TM) i3 CPU M 370 @ 2.40GHz

Not sure whether that is duo core or not. It is about four years old.

I downloaded all the projects (too obsessive) but the specific problem I had was completing Asteroids@Home and moving onto Cosmology@Home.

Thanks again. I will try to implement the changes you recommend.

PS - just googled the Intel Core i3 processor and apparently they are all Duo Cores.

You can open Windows Task Manager and Performance tab. If there is 4 CPUs then Hyper Threading is enabled and BOINC can see 4 cores.
--> You could try value 25%

If Performance tab shows only 2 CPUs then you might try that:
http://windows7themes.net/en-us/how-to-enable-hyperthreading-in-windows-7/

If there is no option to choose "4", then you would need to boot into BIOS and enable Hyper Threading. Here is a procedure to try (answer by user "mgcul8r"):

http://answers.microsoft.com/en-us/windows/forum/windows_7-performance/i-need-to-find-out-how-to-enable-hyperthreading-in/20eb6eec-3d48-405d-b2e4-137a2c7230b5?tab=question&status=AllReplies#tabs

Hold on, you're telling Eddie to go add Hyper Threading (if he doesn't have it already), to then tell him how to use 1 CPU core only for his one task run that he asks for? Why not leave it at 2 cores and go from there? Making it more difficult for nothing?

@Eddie, you say you do not see BOINC switching from one task to the next when one finishes and the next should start. Just a bit of log from my own system, where I only run Seti on my GPU, and then one task at a time. Auto-switch when one task finishes to another task starting.

26/04/2015 14:45:53 | SETI@home | [checkpoint] result 24my12ab.3497.15200.438086664205.12.40_0 checkpointed
26/04/2015 14:47:30 | SETI@home | [checkpoint] result 24my12ab.3497.15200.438086664205.12.40_0 checkpointed
26/04/2015 14:49:00 | SETI@home | [checkpoint] result 24my12ab.3497.15200.438086664205.12.40_0 checkpointed
26/04/2015 14:50:31 | SETI@home | [checkpoint] result 24my12ab.3497.15200.438086664205.12.40_0 checkpointed
26/04/2015 14:52:08 | SETI@home | [task] Process for 24my12ab.3497.15200.438086664205.12.40_0 exited, exit code 0, task state 1
26/04/2015 14:52:08 | SETI@home | Computation for task 24my12ab.3497.15200.438086664205.12.40_0 finished
26/04/2015 14:52:08 | SETI@home | Starting task 28se12ac.9046.13666.438086664197.12.50_0
26/04/2015 14:53:48 | SETI@home | [checkpoint] result 28se12ac.9046.13666.438086664197.12.50_0 checkpointed

As you can see, works without a flaw. Has been doing it like this for absolute ages. So then when it doesn't do that on your system, I would want to know, what is your setup like? Did you suspend all other tasks by chance? Because then yes, you will have to manually resume the next task. BOINC won't do that automatically.

Only tasks showing in the list as status "Ready to start" (has never been started before) or "Ready to run" (has been started at least once, but since auto-paused) will be able to run. All tasks set to Suspended (by whatever, user, waiting for memory, or other) will not run automatically without user intervention.

You're also saying that you feel BOINC Manager's been rushed. BOINC Manager is the graphical user interface, it shows all that the BOINC client is doing, and allows you to easily command it without having to go to the command line and using the BOINC Command Line Tool BOINCCMD for that. The BOINC client (boinc.exe) is what does all the scheduling of what work should run next, what project should run next, what project should download work next, queuing of this work, figure out how long the work takes when run on application X, and keep in mind what the user wants and if that doesn't interfere too much with day to day operations. It does a lot more, but let's simplify things.

This is why we always ask for the start-up messages. It gives a real treasure trove of information, including how many CPU cores BOINC sees your system has, and e.g. when using local preferences, how many cores you've detailed it should use. So always include about the first 40 lines of these messages.

They'll go from Starting BOINC client version 7.5.0 for windows_x86_64 (I'm testing the new one, which is 7.5.0) to Not using a proxy (as I am not using a proxy, if you are it's two lines more).

---

Hold on, you're telling Eddie to go add Hyper Threading (if he doesn't have it already), to then tell him how to use 1 CPU core only for his one task run that he asks for? Why not leave it at 2 cores and go from there? Making it more difficult for nothing?

You are right. Sorry. I wasn't thinking enough :D I had somekind of "maximize all" mode in my head going on. Bigger amount of "cores"/threads the better for Eddie.... but yeah, it isn't like that. He doesn't have to change the HT setting from what it is at the moment.

Thanks guys.

I downloaded most of the projects and then when I clicked "run always" they all started downloading at once. This is what I wanted, but then I saw my computer heating up and the fan going crazy so I thought I would reduce the number of tasks. I suspended most of them except for about five which I had already made some progress on.

I then wanted to only do one at a time, so I suspended the remaining four and waited for the first one to finish. My worry is that with projects like Seti it downloads four or five tasks at a time, so in order to keep the transition smooth I will have to be playing with the suspend and resume buttons, unless I want to do four or five at a time (if doing Seti).

I know it seems like a bit of a pedantic problem, but it just doesn't seem like much to ask for a one at a time task queue.

Ageless, how did you get the auto-switch feauture working for Seti? All my Seti tasks started downloading at the same time.

Thanks

I downloaded most of the projects and then when I clicked "run always" they all started downloading at once.

Downloading work is transferring data from a server onto your computer. It isn't regulated with the "Run X for CPU" options in the Activity menu, but instead with the "Network X" options in that same menu.

When a project's science application is doing calculations on the data in a task, it isn't downloading work, but instead as said doing calculations, using the CPU and where possible, the GPU.

This is what I wanted, but then I saw my computer heating up and the fan going crazy so I thought I would reduce the number of tasks.

Doing calculations on the CPU will indeed heat up the CPU and increase the use of the fan.

I suspended most of them except for about five which I had already made some progress on.

I then wanted to only do one at a time, so I suspended the remaining four and waited for the first one to finish.

And by doing so, you are deciding which task runs next, not BOINC. So no wonder that you don't see BOINC automatically switch to the next task, when you've suspended all but one of them.

My worry is that with projects like Seti it downloads four or five tasks at a time, so in order to keep the transition smooth I will have to be playing with the suspend and resume buttons, unless I want to do four or five at a time (if doing Seti).

You can tell BOINC to download a minute amount of work by setting the work request (Store at least N days of work) plus the additional both to zero. This may still get more than one task in, but then that's what BOINC is made for, storing of work and scheduling when to run all that before the deadline.

You can also tell BOINC to use only one CPU (core), with the "Use at most N% of the CPUs, in your case 50%.

And you can tell BOINC to throttle where necessary, by using the "Use at most N% of CPU time". This value will switch between suspending and resuming of a task, thereby lowering the temperature that the CPU reaches. It will not make the CPU run at a N% load when checked in Windows Task Manager->Processes. If you want something like that, you need a third party app, such as TThrottle.

Ageless, how did you get the auto-switch feauture working for Seti?

I don't manually suspend any of the tasks I have in cache.
I also only have 1 GPU that I run work on, I've told Seti it cannot use my CPU, and I am not using an app_config.xml file with which I tell the GPU to run more than one task on it, which I could if I wanted to.

You have many options to manipulate the amount of work you want to run at the same time. Yet still you ignore my request for posting your start-up messages.

---

Jord, I am sorry but I am not ignoring anything. I don't know what a start up message is. Let me see.

Ageless, do you mean this:

26/04/2015 16:27:06 | | cc_config.xml not found - using defaults
26/04/2015 16:27:06 | | Starting BOINC client version 7.4.42 for windows_x86_64
26/04/2015 16:27:06 | | log flags: file_xfer, sched_ops, task
26/04/2015 16:27:06 | | Libraries: libcurl/7.39.0 OpenSSL/1.0.1j zlib/1.2.8
26/04/2015 16:27:06 | | Data directory: C:\ProgramData\BOINC
26/04/2015 16:27:06 | | Running under account currys
26/04/2015 16:27:06 | | No usable GPUs found
26/04/2015 16:27:06 | | Host name: currys-HP
26/04/2015 16:27:06 | | Processor: 4 GenuineIntel Intel(R) Core(TM) i3 CPU M 370 @ 2.40GHz [Family 6 Model 37 Stepping 5]
26/04/2015 16:27:06 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 sse4_2 popcnt syscall nx lm vmx tm2 pbe
26/04/2015 16:27:06 | | OS: Microsoft Windows 7: Home Premium x64 Edition, Service Pack 1, (06.01.7601.00)
26/04/2015 16:27:06 | | Memory: 3.80 GB physical, 7.60 GB virtual
26/04/2015 16:27:06 | | Disk: 445.49 GB total, 351.43 GB free
26/04/2015 16:27:06 | | Local time is UTC +1 hours
26/04/2015 16:27:06 | | VirtualBox version: 4.3.12
26/04/2015 16:27:06 | Constellation | URL http://aerospaceresearch.net/constellation/; Computer ID 127423; resource share 100
26/04/2015 16:27:06 | Asteroids@home | URL http://asteroidsathome.net/boinc/; Computer ID 167487; resource share 100
26/04/2015 16:27:06 | ATLAS@home | URL http://atlasathome.cern.ch/; Computer ID 19077; resource share 100
26/04/2015 16:27:06 | rosetta@home | URL http://boinc.bakerlab.org/rosetta/; Computer ID 2271548; resource share 100
26/04/2015 16:27:06 | Poem@Home | URL http://boinc.fzk.de/poem/; Computer ID 855448; resource share 100
26/04/2015 16:27:06 | Leiden Classical | URL http://boinc.gorlaeus.net/; Computer ID 129886; resource share 100
26/04/2015 16:27:06 | The Lattice Project | URL http://boinc.umiacs.umd.edu/; Computer ID 117596; resource share 100
26/04/2015 16:27:06 | CAS@home | URL http://casathome.ihep.ac.cn/; Computer ID 48078; resource share 100
26/04/2015 16:27:06 | climateprediction.net | URL http://climateprediction.net/; Computer ID 1361684; resource share 100
26/04/2015 16:27:06 | DENIS@home | URL http://denis.usj.es/denisathome/; Computer ID 1626; resource share 100
26/04/2015 16:27:06 | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 11862226; resource share 100
26/04/2015 16:27:06 | NFS@Home | URL http://escatter11.fullerton.edu/nfs/; Computer ID 1668805; resource share 100
26/04/2015 16:27:06 | FiND@Home | URL http://findah.ucd.ie/; Computer ID 98028; resource share 100
26/04/2015 16:27:06 | Gerasim@home | URL http://gerasim.boinc.ru/; Computer ID 22484; resource share 100
26/04/2015 16:27:06 | EDGeS@Home | URL http://home.edges-grid.eu/home/; Computer ID 46802; resource share 100
26/04/2015 16:27:06 | VirtualLHC@home | URL http://lhcathome2.cern.ch/vLHCathome/; Computer ID 81486; resource share 100
26/04/2015 16:27:06 | LHC@home 1.0 | URL http://lhcathomeclassic.cern.ch/sixtrack/; Computer ID 10358693; resource share 100
26/04/2015 16:27:06 | Milkyway@Home | URL http://milkyway.cs.rpi.edu/milkyway/; Computer ID 617289; resource share 100
26/04/2015 16:27:06 | MindModeling@Beta | URL http://mindmodeling.org/; Computer ID 65178; resource share 100
26/04/2015 16:27:06 | NumberFields@home | URL http://numberfields.asu.edu/NumberFields/; Computer ID 23154; resource share 100
26/04/2015 16:27:06 | pogs | URL http://pogs.theskynet.org/pogs/; Computer ID 479074; resource share 100
26/04/2015 16:27:06 | ibercivis | URL http://registro.ibercivis.es/; Computer ID 424962; resource share 100
26/04/2015 16:27:06 | SAT@home | URL http://sat.isa.ru/pdsat/; Computer ID 30761; resource share 100
26/04/2015 16:27:06 | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 7553279; resource share 100
26/04/2015 16:27:06 | SZTAKI Desktop Grid | URL http://szdg.lpds.sztaki.hu/szdg/; Computer ID 360735; resource share 100
26/04/2015 16:27:06 | Universe@Home | URL http://universeathome.pl/universe/; Computer ID 3171; resource share 100
26/04/2015 16:27:06 | Citizen Science Grid | URL http://volunteer.cs.und.edu/csg/; Computer ID 8604; resource share 100
26/04/2015 16:27:06 | Cosmology@Home | URL http://www.cosmologyathome.org/; Computer ID 268684; resource share 100
26/04/2015 16:27:06 | GPUGRID | URL http://www.gpugrid.net/; Computer ID 209364; resource share 100
26/04/2015 16:27:06 | malariacontrol.net | URL http://www.malariacontrol.net/; Computer ID 1509236; resource share 100
26/04/2015 16:27:06 | RNA World | URL http://www.rnaworld.de/rnaworld/; Computer ID 38726; resource share 100
26/04/2015 16:27:06 | | No general preferences found - using defaults
26/04/2015 16:27:06 | | Reading preferences override file
26/04/2015 16:27:06 | | Preferences:
26/04/2015 16:27:06 | | max memory usage when active: 1946.93MB
26/04/2015 16:27:06 | | max memory usage when idle: 3504.47MB
26/04/2015 16:27:06 | | max disk usage: 100.00GB
26/04/2015 16:27:06 | | don't compute while active
26/04/2015 16:27:06 | | don't use GPU while active
26/04/2015 16:27:06 | | suspend work if non-BOINC CPU load exceeds 25%
26/04/2015 16:27:06 | | (to change preferences, visit a project web site or select Preferences in the Manager)
26/04/2015 16:27:06 | | Not using a proxy

Regards

26/04/2015 16:27:06 | | Processor: 4 GenuineIntel Intel(R) Core(TM) i3 CPU M 370 @ 2.40GHz [Family 6 Model 37 Stepping 5]

I see BOINC sees 4 cores, so your i3 has hyperthreading on.
If you want to run with only 1 core, you'll have to tell BOINC to use 25% at "Use at most N% of the CPUs".

26/04/2015 16:27:06 | | Reading preferences override file

And since you're running with the local preferences, it's best to set the CPU use in there, these preferences override any of the same from the project's web site.
(Tools->Computing preferences->CPU usage)

Mind posting the contents of your global_prefs.xml and global_prefs_override.xml files as well? You can find these in your data directory, C:\ProgramData\BOINC
This is a hidden directory, so to enter there, either fill in the path directory into Windows Explorer, or tell Windows to show hidden files and folders.

You can open all BOINC XML files with Notepad. No special XML editor needed.

---

Hi Jord, I have searched my computer after selecting "show hidden files, folders, and drives". I can only find the global prefs override file, not the global prefs file?

The contents of global prefs override are as follows:

<?xml version="1.0"?>

-<global_preferences>

<run_on_batteries>0</run_on_batteries>

<run_if_user_active>0</run_if_user_active>

<run_gpu_if_user_active>0</run_gpu_if_user_active>

<idle_time_to_run>1.000000</idle_time_to_run>

<suspend_cpu_usage>25.000000</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>0.100000</work_buf_min_days>

<work_buf_additional_days>0.500000</work_buf_additional_days>

<max_ncpus_pct>0.000000</max_ncpus_pct>

<cpu_scheduling_period_minutes>60.000000</cpu_scheduling_period_minutes>

<disk_interval>60.000000</disk_interval>

<disk_max_used_gb>100.000000</disk_max_used_gb>

<disk_max_used_pct>90.000000</disk_max_used_pct>

<disk_min_free_gb>0.100000</disk_min_free_gb>

<vm_max_used_pct>75.000000</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>100.000000</cpu_usage_limit>

<daily_xfer_limit_mb>0.000000</daily_xfer_limit_mb>

<daily_xfer_period_days>0</daily_xfer_period_days>

</global_preferences>

<work_buf_min_days>0.100000</work_buf_min_days>

<work_buf_additional_days>0.500000</work_buf_additional_days>

<max_ncpus_pct>0.000000</max_ncpus_pct>

Those are the ones of interest to me in this case. You're telling BOINC to download enough work for 0.6 days and to contact the projects you allow to fetch work from once that cache amount reaches 0.1 days. it then tops the cache back up to 0.6 days worth of work.
If you really want as low a cache as possible, set both values to zero.

You've also told BOINC to use all CPU cores. 0% is the same as 100%.
Like I said before, if you only want to run one task at the same time, set this to 25%. The value is an integer, which means that when you set 1-24% no CPU cores are used, 25-49% is one core, 50-74% is two cores, 75-99% is three cores and 100% (and 0%) is four cores.

---

Hi Jord, so if I set it so that I only use 25% of my CPU then it will only do one task at a time? So I can just fetch the tasks and not have to worry about it doing four or five at a time or messing about with suspend and resume?

That looks like what I want to achieve. Thanks for that.

In order to save people's time in future would it be possible to create this as an easier to understand feature? I know people are using different computers, but there must be a more user friendly way of doing this?

I will follow your instructions anyway. Thanks for your help, Richie and Jord!

Hi I have changed "on multi processors use at most 0% of the processors" to 25%. So that should be it done! If I want to use two I imagine I just change it to 50%. How come at times in the past I have been able to process more than four tasks at a time? I am sure there were five or six?

I have worked on 7 projects in total. I think with Seti it kept doing bits of different ones. Is the most it can do at one time, four?

Regards