Posts by Nick Name

21) Message boards : Questions and problems : Restrict VirtualBox tasks when PC is in use? (Message 98316)
Posted 4 May 2020 by Nick Name
Post:
Not in a simple way that wouldn't also stop normal non-VB BOINC work; if that's ok then just set BOINC to run when the computer is idle. Otherwise, I would try lowering the number of threads BOINC can use and see if that helps. For example if BOINC is using 100% of an 8-thread machine, set it to 99% instead. If the VB app is multi-threaded you might be able to limit how many threads it uses and that might help. Check with the project(s) on that, you might need to create an app_config file and you'll need the correct names and settings.
22) Message boards : Questions and problems : setup virtualbox (Message 98160)
Posted 29 Apr 2020 by Nick Name
Post:
I don't know if the service installation prevents VirtualBox work from running, but there is no GPU work for any project using VB that I'm aware of.

BOINC@TACC has no work right now and the work supply in general has been very sparse. Right now you'll need to try another project to see if it's actually working.
23) Message boards : Questions and problems : Problems limiting CPU usage on headless linux box (Message 98063)
Posted 25 Apr 2020 by Nick Name
Post:
Great, glad I could help and it was an easy fix.
24) Message boards : Questions and problems : Problems limiting CPU usage on headless linux box (Message 98019)
Posted 23 Apr 2020 by Nick Name
Post:
Hi all. I'm trying to limit cpu usage to half of the available CPUs on a couple of larger hyperthreaded Linux boxes.

In the BOINC directory, I've created a global_prefs_override.xml:

<global_preferences>
<cpu_usage_limit>50.0</cpu_usage_limit>
<max_ncpus_pct>50.0</max_ncpus.pct>
</global_preferences>


But on a 32-thread box, BOINC is still spawning 32 processes. Likewise on a 56-thread server (spawning 56 processes).

Any suggestions?

If this is the actual text, there's a typo.

<max_ncpus_pct>50.0</max_ncpus.pct>
should be
<max_ncpus_pct>50.0</max_ncpus_pct>
25) Message boards : Questions and problems : Bug - bad at meeting deadline - leaves until the very last minute (Message 97707)
Posted 15 Apr 2020 by Nick Name
Post:
BOINC can't predict the future, so if the computer's switched off unexpectedly, how do you want BOINC to know that?

No, but it could actually honor FIFO (First In First Out) and run tasks to completion once they've started, which is the way most people would expect it to work and rightfully so. Doing those two things would solve this for most cases.

I saw this exact scenario many times over the years, it's pretty frustrating to see BOINC letting work sit that you know it can't complete before the deadline, because it's running something else that has a later deadline! It's why I usually run a very low cache, while BOINC still runs things out of order sometimes at least it will have time to get everything returned in time.


Maybe true for most cases. If one of the projects has tasks lasting several weeks as CPDN does at least on slower computers, FIFO would then stop shorter tasks with much shorter deadlines ever running.

The way I see it, whatever system is used, it isn't going to please all of the people all of the time. Personally, I am happy the way it is

I agree, except I would say that projects with extreme deadlines like that should be run by themselves. Work that takes weeks is an outlier and should be handled accordingly. To me, that's more of a project problem than a BOINC problem since most work takes a few hours at most. You should expect problems if you're asking BOINC to do handle project work schedules that are out of the norm, and weeks-long work is definitely that.

That doesn't even apply to the situation being discussed. If the task really sat around for two weeks there was more than enough time to get it processed. Like I said, I've seen this sort of thing for years, I doubt it's going to get fixed. I've found my own way to cope so it's no longer a major problem for me, just an annoyance.
26) Message boards : Questions and problems : Bug - bad at meeting deadline - leaves until the very last minute (Message 97668)
Posted 15 Apr 2020 by Nick Name
Post:
BOINC can't predict the future, so if the computer's switched off unexpectedly, how do you want BOINC to know that?

No, but it could actually honor FIFO (First In First Out) and run tasks to completion once they've started, which is the way most people would expect it to work and rightfully so. Doing those two things would solve this for most cases.

I saw this exact scenario many times over the years, it's pretty frustrating to see BOINC letting work sit that you know it can't complete before the deadline, because it's running something else that has a later deadline! It's why I usually run a very low cache, while BOINC still runs things out of order sometimes at least it will have time to get everything returned in time.
27) Message boards : Projects : Coronavirus projects? (Message 97545)
Posted 13 Apr 2020 by Nick Name
Post:
A list of the projects we're aware of is here.
Folding@Home is pushing a lot of COVID-19 work now if you can get it. Their servers are overwhelmed by all the compute power that's come online in the last few days. That's causing problems both getting and returning work, at least for GPU clients. Work assignment is also buggy, the client can get hung on a server that's not responding or issuing work and won't move on to another. I've set up exclusions in BOINC so that if my machine can't get FAH work, BOINC will resume. If you can put up with these quirks the work has run without any problems for me.


Another non Boinc App which startet a COVID-19 reasearch is DreamLab its available for Android and IOS

Thank you very much. I think someone mentioned that to me on Twitter but I forgot about it. I've added it to the list. Speaking of the list, it's getting a bit cluttered. Time to fix that.
28) Message boards : Questions and problems : BOINC client not detecting VirtualBox installation (Message 97432)
Posted 9 Apr 2020 by Nick Name
Post:
You can check client_state just to be sure it didn't get borked somehow.

* Did you try to crunch Projects using VMs in the past while VT-X / AMD-V / VIA-VT was not enabled? Could be that BOINC has kept this in mind!

To check and fix this, first exit BOINC and make sure, all BOINC-Tasks have really finished.

In your BOINC_Data-Directory you will find a client_state.xml. Open this with a simple editor and search for:

<p_vm_extensions_disabled>1</p_vm_extensions_disabled>


If this is absent or the number is 0 / zero than all is fine. Otherwise change it to 0 / zero <p_vm_extensions_disabled>0</p_vm_extensions_disabled> and safe the file. Be carefull to save it as a real ascii-file

Be carefull that you closed your BOINC-Client successfully before you change anything in client_state.xml. Otherwise BOINC will overwrite your changes
29) Message boards : Projects : Coronavirus projects? (Message 97304)
Posted 6 Apr 2020 by Nick Name
Post:
I...On a side note, Folding at home only fights COVID 19 on CPU....

According to the FAQ, GPU and CPU work is available, although my experience with GPU work it's almost impossible to get.

I'm maintaining a current list of projects with COVID-19 work here. Less publicized projects include Ralph@Home (test project for Rosetta) and BOINC@TACC. Work for these two is spotty at best and TACC has its own problems, but they're on the list if you want to try them.
30) Message boards : Projects : Rosetta or Folding@home? (Message 96894)
Posted 18 Mar 2020 by Nick Name
Post:
Why not do both, Rosetta on CPU and FAH on GPU? At the present time though it appears FAH is out of work so you might only have Rosetta work available. FAH has been overwhelmed by so many people signing up.
31) Message boards : Projects : Coronavirus projects? (Message 96846)
Posted 17 Mar 2020 by Nick Name
Post:
A list of the projects we're aware of is here.
Folding@Home is pushing a lot of COVID-19 work now if you can get it. Their servers are overwhelmed by all the compute power that's come online in the last few days. That's causing problems both getting and returning work, at least for GPU clients. Work assignment is also buggy, the client can get hung on a server that's not responding or issuing work and won't move on to another. I've set up exclusions in BOINC so that if my machine can't get FAH work, BOINC will resume. If you can put up with these quirks the work has run without any problems for me.
32) Message boards : Questions and problems : How would "Resource Share" number work in this case? (Message 96845)
Posted 17 Mar 2020 by Nick Name
Post:
WHAT!! MW gpu apps work fine here on RTX cards. Always has.

I'm guessing a mix-up between Asteroids which doesn't work on RTX, and MilkyWay which does as far as I know.

re. the question about resource share, it shouldn't matter as long as it's greater than zero, but I would just leave them equal. You can always adjust later if you add more projects.
33) Message boards : BOINC Manager : Manual suspend keeps getting reset (Message 96261)
Posted 2 Mar 2020 by Nick Name
Post:
Or suspend a task, after which Boinc decides to load more tasks from that same project.
No. Work fetch from a given project is inhibited if a task from that project is suspended.

You mean, if the project is suspended.
If there's one paused task in queue, the client still redownloads more, until it has enough tasks to satisfy it's daily task buffer (eg preload tasks for 1 day, and there'll be a whole list of tasks it will download).

I haven't seen that behavior, however I have seen this, many times:

3/2/2020 12:32:49 AM | Amicable Numbers | Sending scheduler request: To report completed tasks.
3/2/2020 12:32:49 AM | Amicable Numbers | Reporting 1 completed tasks
3/2/2020 12:32:49 AM | Amicable Numbers | Not requesting tasks: some task is suspended via Manager
3/2/2020 12:32:51 AM | Amicable Numbers | Scheduler request completed

Richard is correct, and this has been how BOINC works for as long as I can remember.
34) Message boards : GPUs : Big ask - auto throttling of CPU tasks when using GPUs? (Message 96246)
Posted 1 Mar 2020 by Nick Name
Post:
There's no way to "automatically" do what you want or do any dynamic load balancing, at least within BOINC itself. The easiest thing to do is lower your CPU usage limit. BOINC will typically over-commit the CPU when running GPU work, e.g. if BOINC is set to use four CPU threads and you have two GPUs BOINC will run six jobs in total. I take this into account and set CPU use so the number of jobs doesn't exceed the CPU thread count.
35) Message boards : GPUs : PCI express risers to use multiple GPUs on one motherboard - not detecting card? (Message 96245)
Posted 1 Mar 2020 by Nick Name
Post:
One, because the higher end GPUs don't need to run at their rated power to do the tasks.
Two, because higher end GPUs have more cores.
Three, because their performance is nearly identical (with overclock) at 20-60% power cap, vs stock.
Running at stock wattage, the GPU frequency is limited. By cutting power and overclocking, you can get close to this performance, at a fraction of the power. If you run 3 GPUs, capped at 35-40% power cap, you can add a fourth one and still use the same amount of power at the wall, but with 25% extra performance.
And four, doubling up on Einstein jobs on a high end GPU, is an easy way to fully utilize the GPU.

Using a smaller GPU will perform a lot slower. For instance, the 2080Ti is about twice as fast as a 2060 on low intensity tasks. The 2060 uses 125W out of 170W stock, the 2080 Ti does it at 150W. Multiply the 125W by 2, to get similar performance as the 2080 Ti, and it's easy to see how the 2080Ti is much more efficient (at similar performance levels).

For high intensity tasks, or two tasks per GPU, a 2060 will run at 140-150W, while a 2080 Ti does it at 190-220W.
Usually high intensity tasks also have qrb (quick return bonus), which seem to favor higher end GPUs over multiple lower speed ones.

I'll have to revisit this as the last time I spent any significant time testing this was on Pascal 1080 and this kind of performance wasn't possible. Performance started to degrade more than I was willing to tolerate around an 80% power limit and I was not able to recover any by overclocking. Even a 10 mhz overclock at that limit sent invalids skyrocketing. I have limited a Radeon VII to this extent but there I have control over voltage which helped stabilize things.
36) Message boards : GPUs : PCI express risers to use multiple GPUs on one motherboard - not detecting card? (Message 96227)
Posted 1 Mar 2020 by Nick Name
Post:

I don't run them beyond 225W, as for smaller projects (like Einstein) they would hit past their max rated GPU frequency (sometimes 2200-2400Mhz) and crash the task; or go into limp mode (1350Mhz) and do it very slow.

Why not just buy a lower power card if you're going to limit the power so severely?
37) Message boards : Questions and problems : Questions about running BOINC (Message 96163)
Posted 28 Feb 2020 by Nick Name
Post:
1. I have SETI@Home and World Community Grid as my projects. I have an 8 core CPU and an Nvidia graphics card. Right now, BOINC will use all 8 cores for World Community Grid and the SETI task is using my graphics card and a portion of one core. How do I get a balance to where each project is using 4 cores? Is that even possible?

The only way to get this specific control is to run multiple clients. You would then set each client to only use four cores, and attach one to WCG and the other to SETI. It can be done but it's not really worth it just for this. BOINC will balance the two pretty well over time, but think in terms of days or weeks, not hours.

2. Is it faster for BOINC to process tasks on a Linux-based OS or Windows 10? If a Linux-based OS, is there a preference?

BOINC Manager 7.14.2
Windows 10
I've said the projects above.
I've installed BOINC as a service.

I"ve seen little difference in CPU work, but the latest AMD CPUs with higher core counts might show a significant difference. There can be a substantial difference in GPU work due to Windows WDDM which adds significant overhead. SETI is a bit unusual in that there is a special app that only runs on Nivida cards under Linux, it's orders of magnitude faster than anything on Windows. The gain is partly due to Linux advantages and partly due to highly optimized programming.

It's interesting you say you are running GPU work with BOINC installed as a service. My understanding is that hasn't worked since BOINC version 6.
38) Message boards : Questions and problems : Hi, I'm new to BOINC. I have some question from my noob perspective. (Message 96052)
Posted 25 Feb 2020 by Nick Name
Post:
1) What do you think BOINC and blockchain(or cryptocurrency) have in common?

Nothing directly.
2) I read about VC from wikipedia and BOINC from the article(BOINC: A Platform for Volunteer Computing), and I found that there are no actual merits for the participants who provide their computing device.

"Actual merits" depends on what you mean. Recognition for contributions to research does exist in some projects, e.g. GPUGrid lists publications your computer(s) has contributed to. Most projects give badges based on your contribution, how much value you think that has depends on your perspective.
2.1 The existing commercial clouds maintain and provide their computing power for money.

This is actually partly why BOINC exists. Many projects are short on money and BOINC provides a way to access pretty powerful computing resources at relatively low cost.
2.2 The blockchain participants participate for rewards(coin)
Then why do the VC participants provide their computing devices? I'd really like to know.
Are they just enjoying donating or is there any kind of reward that I don't know?
Thank you.

Gridcoin, which uses blockchain technology, was built on top of BOINC and does provide a - these days very small - financial reward for running BOINC. You can look into the details on it further on your own if you wish. For most of those folks obviously the potential monetary gain is the primary motivation for running BOINC. For the rest of us, it's more or less a hobby but one we find worthwhile. I think most participants have interests in both technology and science, and BOINC is a neat way to combine the two. The team component has added further interest as competitions and challenges have been created over the years and a lot of people really enjoy that. In short, what you're asking is really no different than why people build hot rods, do woodcraft or anything else. BOINC however actually might help to change the world one day. The ultimate reward for many of us is helping advance science for the benefit of humanity.
39) Message boards : Questions and problems : Manager without client (Message 95849)
Posted 14 Feb 2020 by Nick Name
Post:
The only Android client I'm aware of that allowed remote management was NativeBOINC. Unfortunately development stopped several years ago. I have several ARM boards myself that are idle partly because of this problem.
40) Message boards : Questions and problems : Manager without client (Message 95797)
Posted 11 Feb 2020 by Nick Name
Post:
You can control remote clients with the standard manager.

https://boinc.berkeley.edu/wiki/Controlling_BOINC_remotely

You can only control one at a time. BOINCTasks is designed to show/manage multiple clients together, which is why it's a better option if you need to manage multiple clients. You do have to install WINE if you want to run it on a Linux machine as there's not a native Linux app. You can use it on a Windows machine to control Linux clients, no problem.

I'd forgotten about it but there is also boinctui for Debian distros.


Previous 20 · Next 20

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.