Hi,

i've configured Boinc to use at most 75% of CPU time to prevent heating and throttling the processor. The problem is, Amicable doesn't respect those limits.. it uses 100% CPU time no matter what. In the past i was able to run 2 Amicable task at once without any problems... but now it's not feasible as CPU temps go skyrocket.

Is this the expected behaviour?. Is there something i've misconfigured?

BOINC client version 7.2.42 for x86_64-pc-linux-gnu
Processor: 8 GenuineIntel Intel(R) Core(TM) i7-4790 CPU @ 3.60GHz [Family 6 Model 60 Stepping 3]
OS: Linux: 4.4.104-39-default
Memory: 7.70 GB physical, 2.01 GB virtual
VirtualBox version: 5.1.28r117968

Regards

It's better to set the number of CPU cores to use and run only one task at a time. Try setting 2, 3, 4 cores to find which setting doesn't overheat your CPU. Setting the % of CPU time to use sometimes gives an invalid result for WU, and I have no idea why, so I had to turn off this setting.

P.S. Actually, it shouldn't overheat when running Amicable. Amicable uses only integer arithmetic, no heavy SSE/AVX instructions at all. Check you CPU cooler.

Hi,
i'll try lowering # jobs & increasing cores, otherwise boinc will have more slots available to other projects.

This cooler is a piece of shit, but is clean & thermal compound is fine.. ambient is a little bit high here at 30ºC, but with all other projects running.. is fine @70-75ºC .
With Amicable running along with others i got 88-92ºC.. after a few minutes.. .anyway, i'll try your advice.. will begin from low 1... going up to test the limits.
Thanks

Hello,
Also the same for me. the WU use 100% of CPU.
My BAM settings are set à 75% CPU time.
I run one WU at the time, but it use all cores. So impossible to run other projects at the same time.
If I reduce the available core, it crash.
It not respect the limits. It is the only problem !!!

---

If I reduce the available core, it crash.

This crash is fixed already, try again.

If I reduce the available core, it crash.

This crash is fixed already, try again.

Hello Sergei,
Thank you.
I forgot to also write that I have set a limit of 4 cores in project preferences.
Of course also in computing preferences (75%) and in my BAM (local preferences,
It nothing change. the WU take all cores at 100% load with result of overheat of CPU.
https://sech.me/boinc/Amicable/workunit.php?wuid=6732795.
estimated commputing : 2h30

It seems project not validate the limits of computation. On all other project, all WU run at 75%. Why not here ?
Best regards from the first belgian team admin.

---

You have 4 cores per WU, but you also need to set "Max # jobs" to 1. Otherwise BOINC will start 2 WUs that will take all 8 cores.

P.S. Your CPU shouldn't overheat when running Amicable even at 100% on all cores. It doesn't even use SSE, it's only integer math on Amicable. Check your CPU cooler.

You have 4 cores per WU, but you also need to set "Max # jobs" to 1. Otherwise BOINC will start 2 WUs that will take all 8 cores.

P.S. Your CPU shouldn't overheat when running Amicable even at 100% on all cores. It doesn't even use SSE, it's only integer math on Amicable. Check your CPU cooler.

Hello,
I repeat,
It run only one WU at the time ! At this time, I have only one WU on host. So it can not run two WU
The problem is that project do not respect limitations of 75% of CPU time and so take all ressources of host.

This was already answered in the second post of this thread. CPU time limitation done by BOINC doesn't work well with Amicable, there was a lot of invalid results.

Hello Sergei.
Sorry, I had not read about no limitation.
Little precision,
My hosts runs all 24/24, 7/7,.......
No overclocking.
On all project, there are 8 simultanous CPU WU and one GPU WU.
Temperature never exceed 55°C. Normal with such load.
So it is not a problem of my cooling (Fan artic Freezer 33 : TDP 320W ( for a thermal CPU of 95W)
On this host, I manually reduced CPU speed from 3.4Ghz to 3.0Ghz..
When I run one of your task, I also disable GPU, but temperature increase to 70/75°C !!!
And it take all ressources of host who become almost unusable (except task manager).


To everyone , i have tested watercooling ( Not very efficient on (very high CPU charge,
reaction time is to low.

75°C is normal temperature for CPU, nothing bad will happen to it. Amicable runs at the lowest possible priority, so even if it uses all cores, the system should be responsive.

75°C is normal temperature for CPU, nothing bad will happen to it. Amicable runs at the lowest possible priority, so even if it uses all cores, the system should be responsive.

Hello,
Please check
https://ark.intel.com/products/52214/Intel-Core-i7-2600K-Processor-8M-Cache-up-to-3_80-GHz
TMax in Celsius : 72.6°C !!!
On all my "mid range" mainboard (ASUS) T-Max is 70°C. With result that MB put itself in safe mode.
75°C is not a normal temperature !

On all my hosts, with my special cooler, it never go above 55°C !
With the very hot weather in Europe, sometimes it go to 60°C

I see in task manager it is low priority. But it use 100% of cores and with full load (charge).

Is it possible to reduce manually the CPU charge ?
Best regards

Don't mix Tmax and Tjmax. Tjmax for i7 2600k is 100C. It starts throttling at 90C. 75C is totally safe. If your processor reports 75C, it's a temperature of its internal sensor, temperature at the heat spreader (which must not be higher than Tmax) will be much less than 75C.

Don't mix Tmax and Tjmax. Tjmax for i7 2600k is 100C. It starts throttling at 90C. 75C is totally safe. If your processor reports 75C, it's a temperature of its internal sensor, temperature at the heat spreader (which must not be higher than Tmax) will be much less than 75C.

Dear Sergei.
Thank you for following. Very few Admin do it like you. Congratulations.

I am not sure about 2600K. I also not mix. At 100°C CPU smelt (junction fusion T°).
Intel recomend to not go above 72°C. So I will not let my CPU go above the limit.
If I am located at north (south) pole, no problem. But at end of the year I move to Afirica.
I search the best cooling solution. Be assured.

Then the very hgh CPU temperature increase also the temperature of mainboard !!!
At 60°C ASUS mainboard go on warning mode.
The CPU temperature spread all over inside host.
And be sure, all my host are fully correctly cooled. With correct air flow : one push of 18Cm in front (54CFM)
One 12Cm ( 30CFM) above fan extractor. One rear 9cm extractor. And again inside a 9cm directly on GPU (front to rear direction)
And two 12cm on CPU fan. Of course in same direction !!! I am not stupid (I think)
No, there is no dust on fan !!! At least twice a month, I clean dust.
I will take no any risk to destroy my hosts !!!
This happens only with your project !
Why you not adapt ???
Look my projects, also the projects of my team. We are serious.
Kind and friendly regards from Belgium.
Philippe.

---

On Linux I suggest installing the sysfs package so you can set ignore_nice_load [1]. Unfortunately even a fully niced process will be able to crank the CPU up.

[1] https://www.kernel.org/doc/html/v4.14/admin-guide/pm/cpufreq.html