Come on admit it, you have an E5-2670 crunching on Boinc and/or Seti@Home. How is it doing?

Message boards : Number crunching : Come on admit it, you have an E5-2670 crunching on Boinc and/or Seti@Home. How is it doing?
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 . . . 8 · Next

AuthorMessage
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1884139 - Posted: 16 Aug 2017, 6:33:00 UTC
Last modified: 16 Aug 2017, 6:34:08 UTC

My next experiment with the E5-2670 box is to upgrade to a MS Windows Server 2012 r2 (standard) so I can practice being a system admin and other assorted skills. Since it supports 2 Virtual Machines, I currently am planning on installing a Ubuntu server for additional educational opportunities. I scored some training on both servers on Udemy so I am going to "try" to actually get half way competent. Then I can start more IT consulting. :)

Yes, it will come up running Boinc too.

Tom
A proud member of the OFA (Old Farts Association).
ID: 1884139 · Report as offensive
Marco Vandebergh ( SETI orphan )

Send message
Joined: 27 Aug 10
Posts: 39
Credit: 12,630,994
RAC: 9
Netherlands
Message 1884242 - Posted: 16 Aug 2017, 18:12:41 UTC - in response to Message 1883597.  

I run everything on GPU, 3 tasks at the same time. A guppi task runs approx 8 minutes. Don't really see the benefit from the schedular in my situation.
I'm running a 1080Ti though.


I get the impression that you are running World Computing Grid as your main cpu task(s)?

Thank for sharing your command line parameters.

Apparently my Gtx 1060 (and all others) don't do multiple Seti/SOG work units. They slowdown too much. :(

Do you use automatic "suspend the gpu when computer is in use" for your gaming or what?

What games are you currently enjoying?

Thanks,
Tom


Hi Tom, currently im playing The Witcher 3 only. No i just suspend when i am going to play. Just manually.

No it was SETI as main task before, but the units are so much faster on GPU that i deceided to run all alone SETI on GPU and all WCG on CPU.

I could buy more GPU, but want to wait for Volta, then the Ti version see what that does, 1080Ti will be crunching aside the new one overtime.

Or maybe AMD side, don't really know how they perform with seti. I had 2 x HD6970 before, they were OK at that time.
ID: 1884242 · Report as offensive
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1884401 - Posted: 17 Aug 2017, 6:12:06 UTC - in response to Message 1884242.  
Last modified: 17 Aug 2017, 6:12:31 UTC

No it was SETI as main task before, but the units are so much faster on GPU that i deceided to run all alone SETI on GPU and all WCG on CPU.

I could buy more GPU, but want to wait for Volta, then the Ti version see what that does, 1080Ti will be crunching aside the new one overtime.

Or maybe AMD side, don't really know how they perform with seti. I had 2 x HD6970 before, they were OK at that time.


Apparently you can generate an RAC of something 30,000 on Seti just using the cpu's. And running the Lunatic's AVX cpu app is significantly faster than the Stock application. (30%-40% faster).

I am also running WCG on another i7-3770 box with a gxt 1060 (and it has headed up towards 16,000+ RAC[single task] on just the gpu :) I run a little WCG and Rossetti on my E5 box but I am trying to mostly do Seti while keeping it "busy" through the Tuesday black out, lately with Eisenstein@Home "gpu" only.

I am a gaming newbie but will see if I can find out anything about Witcher 3.

Tom
A proud member of the OFA (Old Farts Association).
ID: 1884401 · Report as offensive
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1884495 - Posted: 17 Aug 2017, 17:07:20 UTC - in response to Message 1884139.  

My next experiment with the E5-2670 box is to upgrade to a MS Windows Server 2012 r2 (standard) so I can practice being a system admin and other assorted skills.


Having now installed MS Windows Server 2012 r2 I have some advice.

Don't install this type of OS [SERVER] unless you have an over-riding need like I did/do. Even when installed with the GUI it feels awkward and clumsy because it is a "Server" not a workstation.

It doesn't even have a shutdown button in the Gui. You have to go to the powershell and type "shutdown /s" w/o the quotes. No, I don't know what the /s means. I am working my way through the very beginning of an "install and configure" the MS Server 2012 r2 course.

I ran across a method of installing MS Security Essentials (which is unsupported) so I don't have to pay even more for a Server level anti-virus. I did manage to get the two profiles merged on the Seti Website. Right now it still looks like I may end up going thru the application test cycle again.

If you change OS's it might be a good idea to immediately suspend get new tasks until you have the profiles merged. Then it MIGHT just send you what it was sending you.

Otherwise it seems to be just munching along like it used to.

Tom
A proud member of the OFA (Old Farts Association).
ID: 1884495 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1884540 - Posted: 17 Aug 2017, 19:36:54 UTC - in response to Message 1884495.  

My next experiment with the E5-2670 box is to upgrade to a MS Windows Server 2012 r2 (standard) so I can practice being a system admin and other assorted skills.


Having now installed MS Windows Server 2012 r2 I have some advice.

Don't install this type of OS [SERVER] unless you have an over-riding need like I did/do. Even when installed with the GUI it feels awkward and clumsy because it is a "Server" not a workstation.

It doesn't even have a shutdown button in the Gui. You have to go to the powershell and type "shutdown /s" w/o the quotes. No, I don't know what the /s means. I am working my way through the very beginning of an "install and configure" the MS Server 2012 r2 course.

I ran across a method of installing MS Security Essentials (which is unsupported) so I don't have to pay even more for a Server level anti-virus. I did manage to get the two profiles merged on the Seti Website. Right now it still looks like I may end up going thru the application test cycle again.

If you change OS's it might be a good idea to immediately suspend get new tasks until you have the profiles merged. Then it MIGHT just send you what it was sending you.

Otherwise it seems to be just munching along like it used to.

Tom

I'm not sure why are are having to run shutdown from the command line. You can press the windows button and then click the power button on the start screen just like in the desktop version of Windows.
If you add the server feature Desktop Experience then you will get most of the things from the desktop OS that you may want for a workstation setup.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1884540 · Report as offensive
Marco Vandebergh ( SETI orphan )

Send message
Joined: 27 Aug 10
Posts: 39
Credit: 12,630,994
RAC: 9
Netherlands
Message 1884889 - Posted: 19 Aug 2017, 11:03:15 UTC - in response to Message 1884540.  
Last modified: 19 Aug 2017, 11:07:38 UTC

Tom, do you know how to set up boinc to get astropulse and multibeam folding together in the same time on GPU?

I did it before with ATI cards i had, and this worked, now for some reason it does not work anymore...

Config.xml with concurrent tasks set does have no effect... On WCG it does have effect when used this way.


I want to fold 3 MB tasks and 1 Astropulse task at the same time.
ID: 1884889 · Report as offensive
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1884896 - Posted: 19 Aug 2017, 12:53:49 UTC - in response to Message 1884540.  

I'm not sure why are are having to run shutdown from the command line. You can press the windows button and then click the power button on the start screen just like in the desktop version of Windows.


Sorry, no such animal showing up when I click the windows button

If you add the server feature Desktop Experience then you will get most of the things from the desktop OS that you may want for a workstation setup.


I will look for that feature when I am a little less tired. Thank you.

Tom
A proud member of the OFA (Old Farts Association).
ID: 1884896 · Report as offensive
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1884900 - Posted: 19 Aug 2017, 13:29:10 UTC - in response to Message 1884889.  
Last modified: 19 Aug 2017, 13:33:30 UTC

Tom, do you know how to set up boinc to get astropulse and multibeam folding together in the same time on GPU?

I did it before with ATI cards i had, and this worked, now for some reason it does not work anymore...

Config.xml with concurrent tasks set does have no effect... On WCG it does have effect when used this way.


I want to fold 3 MB tasks and 1 Astropulse task at the same time.


I see you have a gtx 1080 so it should be possible. I am running a gtx 1060 that will not run more than 1 task at a time without massive performance loss.

Do you have an app_config.xml file in each project file? You can use the "project_max_concurrent" to limit say the WCG project to half the available cores you have on your E5-2670.

I am not sure if the "max_concurrent" command can be limited to only the gpu. I am sure that if you run it w/o the max_concurrent it will run 4 tasks including both the MB and the Astropulse.

With the max_concurrent command outside the <gpu> tags it will limit the total number of cpu tasks executing no matter how many cores you have.

app_config.xml contents
<app_config>
<app>
<name>setiathome_v8</name>
<gpu_versions>
<gpu_usage>0.25</gpu_usage>
<cpu_usage>1.0</cpu_usage>
<max_concurrent>3</max_concurrent>
</gpu_versions>
</app>
<app>
<name>astropulse_v7</name>
<gpu_versions>
<gpu_usage>0.25</gpu_usage>
<cpu_usage>0.50</cpu_usage>
<max_concurrent>1</max_concurrent>
</gpu_versions>
</app>
</app_config>



The following are comments
<gpu_usage>0.5</gpu_usage>
<cpu_usage>0.5</cpu_usage>


The trick is with the gpu_usage. 0.5 for 2 tasks, 0.33 for 3 tasks etc.

There is also a project maximum that you can use <project_max_concurrent>4</project_max_concurrent> that will limit the total number of tasks for the cpu. Possibly for the gpu. Put in the last line before the </app_config>.

I raised this issue in another thread where I got the advice for limiting the total number of apps running on the cpus. The gpu_usage is common for setting up multiple tasks on a gpu.

For SOG gpu tasks you want 1 core / task. Apparently for astropulse you can run 2 tasks on a cpu to drive 2 astropulse tasks.

I lack experience to tell you if you can limit the number of gpu tasks / project in another .xml file.

I have not tested the max_concurrent on my gpu since it doesn't multi-task worth a hoot.

HTH,
Tom
A proud member of the OFA (Old Farts Association).
ID: 1884900 · Report as offensive
Marco Vandebergh ( SETI orphan )

Send message
Joined: 27 Aug 10
Posts: 39
Credit: 12,630,994
RAC: 9
Netherlands
Message 1884916 - Posted: 19 Aug 2017, 15:08:47 UTC - in response to Message 1884900.  

Thanks Tom, this works perfect, just copied your code.

I let it fold for a while now.

Problem was, because i do not run 24/7, it only runs the MB units first, eventually the AP units timed out, cumputer did just not crunch it in time. ( this was a time ago before my 1080 )

So now it just crunces also the AP units, they are not many, 3 atm. But when they arrive they won't get lost now.

Thanks!
ID: 1884916 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1884917 - Posted: 19 Aug 2017, 15:08:50 UTC - in response to Message 1884889.  

Very late to this party.

Best efficency for a 1080Ti is 3 work units of either type, ie.. 3 MB (SoG) or 2 MB(SoG)+1 AP or 1 MB+2AP

3 MB(SoG) +1 AP would decrease the efficency of that GPU. What I would do is set your app_config.xml to run either 3 MB and/or 3 AP at the same time on the GPU.

This can easily be done by the following.
<app_config>
<app_version>
<app_name>setiathome_v8</app_name>
<plan_class>opencl_nvidia_SoG</plan_class>
<avg_ncpus>1</avg_ncpus>
<ngpus>0.33</ngpus>
<cmdline>-sbs 1024 -period_iterations_num 1 -hp -high_prec_timer -high_perf -tt 1500</cmdline>
</app_version>
<app_version>
<app_name>astropulse_v7</app_name>
<plan_class>opencl_nvidia_100</plan_class>
<avg_ncpus>1</avg_ncpus>
<ngpus>0.33</ngpus>
<cmdline>-unroll 28 -oclFFT_plan 256 16 256 -ffa_block 12288 -ffa_block_fetch 6144 -tune 1 64 4 1 -tune 2 64 4 1 -hp </cmdline>
</app_version>
</app_config>


This should allow any combination of what I outlined above to work with the 1080Ti. If you want to also run CPU task the you can add<app_config>

<project_max_concurrent>x</project_max_concurrent>

Just before the </app_config> and take into account how many GPU task are running (in this case 3) and add that to the total amount of CPU task you want. But you don't want to overload you CPU too much as it can affect the rate of processing. In your case, you have a 32 core, running 3 GPU (mixed) so you could say

<project_max_concurrent>32</project_max_concurrent>

it would then run 3 GPU tasks and 29 CPU task. (a little overkill) Better to so with a smaller number and leave some cores free.
ID: 1884917 · Report as offensive
Marco Vandebergh ( SETI orphan )

Send message
Joined: 27 Aug 10
Posts: 39
Credit: 12,630,994
RAC: 9
Netherlands
Message 1885011 - Posted: 20 Aug 2017, 11:05:58 UTC - in response to Message 1884917.  

Ok i will give that a try, main reason was, that my system was not folding AP and MB together and the AP become lost due to the many MB units in row.


I now set your config in the XML and give that a go.


Thanks for the help all :)
ID: 1885011 · Report as offensive
Profile ML1
Volunteer moderator
Volunteer tester

Send message
Joined: 25 Nov 01
Posts: 20140
Credit: 7,508,002
RAC: 20
United Kingdom
Message 1885028 - Posted: 20 Aug 2017, 13:45:31 UTC

For interest, anyone got a comparison for running the system with a lean Linux distro compared to the much larger Windows installs?

The top spots look to go to Linux for the time being at least:

Top hosts


Is there a few more percent to be gained there??

Happy fast crunchin',
Martin
See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)
ID: 1885028 · Report as offensive
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1885031 - Posted: 20 Aug 2017, 13:59:43 UTC - in response to Message 1885028.  

For interest, anyone got a comparison for running the system with a lean Linux distro compared to the much larger Windows installs?

The top spots look to go to Linux for the time being at least:

Top hosts


Is there a few more percent to be gained there??

Happy fast crunchin',
Martin


Yes. My impression is you can get from 5%-30% or more production running generic Linux. And once I can nail a "turn key" install of Linux I will be tempted to try it out. The catch is I don't want to do that on my E5 until I am done with my MS Server lessons :)

Tom
A proud member of the OFA (Old Farts Association).
ID: 1885031 · Report as offensive
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1885034 - Posted: 20 Aug 2017, 14:10:39 UTC - in response to Message 1884917.  

This should allow any combination of what I outlined above to work with the 1080Ti.


Neat. I wondered about 4 gpu tasks being too many. Thank you for the solution.

I understand it won't limit the MB tasks and "reserve" a space for the Astropulse tasks but I am not sure my solution would have either.

Is there a way to limit the mix to 2 MB tasks and and 1 astropulse task? And if there are no astropulse tasks, just run two MB tasks?

That was my reading of the original presenting problem (scaled down to 3 from 4 tasks).

Thanks.

Tom
A proud member of the OFA (Old Farts Association).
ID: 1885034 · Report as offensive
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5124
Credit: 276,046,078
RAC: 462
Message 1885036 - Posted: 20 Aug 2017, 14:14:10 UTC - in response to Message 1884917.  



This can easily be done by the following in an app_config.xml file?
<app_config>
  <app_version>
    <app_name>setiathome_v8</app_name>
    <plan_class>opencl_nvidia_SoG</plan_class>
    <avg_ncpus>1</avg_ncpus>
    <ngpus>0.33</ngpus>
    <cmdline>-sbs 1024 -period_iterations_num 1 -hp -high_prec_timer -high_perf -tt 1500</cmdline>
  </app_version>
   <app_version>
    <app_name>astropulse_v7</app_name>
    <plan_class>opencl_nvidia_100</plan_class>
    <avg_ncpus>1</avg_ncpus>
    <ngpus>0.33</ngpus>
    <cmdline>-unroll 28 -oclFFT_plan 256 16 256 -ffa_block 12288 -ffa_block_fetch 6144 -tune 1 64 4 1 -tune 2 64 4 1 -hp </cmdline>
  </app_version>
</app_config>



Or was that a cconfig.xml file?
Tom
A proud member of the OFA (Old Farts Association).
ID: 1885036 · Report as offensive
J. Mileski
Volunteer tester
Avatar

Send message
Joined: 9 Jun 02
Posts: 632
Credit: 172,116,532
RAC: 572
United States
Message 1885037 - Posted: 20 Aug 2017, 14:28:58 UTC - in response to Message 1881064.  

Hi,
As a nearly brand new owner of a 2 cpu E5-2670 Windows box I am trying to find more people with such hardware and see how they are "doing". eg. what kind of RAC, what kind of cpu tasks / day are you experiencing?

I would be interested in the results of Linux boxes too.

I have one person who is apparently cranking out 32,000+ RAC on a box with no gpu processing.

I have had to reinstall from scratch, twice while experimenting with Raid 1 (thou shall backup EVERYTHING next time!) So even with the help of a Gtx 1060 I am still south of 10,000 RAC after two days :)

So what are your(s) doing? Brag a little :D

Thank you,
Tom

I've got 2 E5-2670 0 cpu's in 2 boxes because I bought the wrong motherboard. So my dual cpu xeon is a Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz [Family 6 Model 79 Stepping 1] named martouf hostid=8070937.
The first of my 2 Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz [Family 6 Model 45 Stepping 6] is PC-Office hostid=5403642
The 2nd one has been offline until Tuesday this past week waiting for a new motherboard as I had one burn up on me. anise hostid=5541574

I only have one 250GB SSD in each as they are dedicated crunching machines.
ID: 1885037 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1885038 - Posted: 20 Aug 2017, 14:29:12 UTC
Last modified: 20 Aug 2017, 14:31:23 UTC

Is there a way to limit the mix to 2 MB tasks and and 1 astropulse task? And if there are no astropulse tasks, just run two MB tasks?


Try this

<app_config>
<app_version>
<app_name>setiathome_v8</app_name>
<plan_class>opencl_nvidia_SoG</plan_class>
<avg_ncpus>1</avg_ncpus>
<ngpus>0.33</ngpus>
<cmdline>-sbs 1024 -period_iterations_num 1 -hp -high_prec_timer -high_perf -tt 1500</cmdline>
<max_concurrent>2</max_concurrent>
</app_version>
<app_version>
<app_name>astropulse_v7</app_name>
<plan_class>opencl_nvidia_100</plan_class>
<avg_ncpus>1</avg_ncpus>
<ngpus>0.33</ngpus>
<cmdline>-unroll 28 -oclFFT_plan 256 16 256 -ffa_block 12288 -ffa_block_fetch 6144 -tune 1 64 4 1 -tune 2 64 4 1 -hp </cmdline>
</app_version>
<max_concurrent>1</max_concurrent>
</app_config>

This should only allow 2 MB to run at any time and save a spot for 1 AP if/when it is available. This will only work for a single GPU in a machine. If you were to run 2 GPUs it would only allow 2 total at any time on either GPU, so for what you are requesting for a single GPU it will work, I think. Been a while since I ran a single GPU lol
ID: 1885038 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1885048 - Posted: 20 Aug 2017, 14:59:35 UTC - in response to Message 1885028.  

For interest, anyone got a comparison for running the system with a lean Linux distro compared to the much larger Windows installs?

The top spots look to go to Linux for the time being at least:

Top hosts


Is there a few more percent to be gained there??

Happy fast crunchin',
Martin

Petri's GPU app has significant gains and is currently only for Linux.
For the CPU apps there is no advantage to running any specific OS.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1885048 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1885079 - Posted: 20 Aug 2017, 16:25:46 UTC - in response to Message 1885028.  
Last modified: 20 Aug 2017, 16:26:03 UTC

[quote]

The top spots look to go to Linux for the time being at least:

/quote]

I like how you put "for the time being"...lol

Yes, many of those boxes are about to be displaced further DOWN the list by the end of the week ;)

Zalster
ID: 1885079 · Report as offensive
Marco Vandebergh ( SETI orphan )

Send message
Joined: 27 Aug 10
Posts: 39
Credit: 12,630,994
RAC: 9
Netherlands
Message 1895272 - Posted: 14 Oct 2017, 11:32:29 UTC - in response to Message 1884917.  
Last modified: 14 Oct 2017, 12:10:04 UTC

For some reason, it did work first.

But i upgraded to the latest boinc version, and again no folding together of AP and MB WU's.
Very annoying, especially the given up time i have i would like to contribute to the max. The AP units will get out of time because the system wants to crunch the MB tasks first.

Code in app_config.xml:

<app_config>
<app_version>
<app_name>setiathome_v8</app_name>
<plan_class>opencl_nvidia_SoG</plan_class>
<avg_ncpus>1</avg_ncpus>
<ngpus>0.33</ngpus>
<cmdline>-sbs 1024 -period_iterations_num 1 -hp -high_prec_timer -high_perf -tt 1500</cmdline>
</app_version>
<app_version>
<app_name>astropulse_v7</app_name>
<plan_class>opencl_nvidia_100</plan_class>
<avg_ncpus>1</avg_ncpus>
<ngpus>0.33</ngpus>
<cmdline>-unroll 28 -oclFFT_plan 256 16 256 -ffa_block 12288 -ffa_block_fetch 6144 -tune 1 64 4 1 -tune 2 64 4 1 -hp</cmdline>
</app_version>
</app_config>


Also, if someone would help me to add the CPU AVX app to this config file i can also put my additional 12 cores to fold also on SETI.
I used the latest SoG intaller.
( sorry, im not lazy, its a learning process to me )

I stopped crunching totally on WCGRID last week, because, in my country some expensive medicines ( due to the unacknowledged monopoly positions of the medecine companies probably ) are not available for the "worker class" people ( just like me ). Because they make them too expensive while they can. And i do NOT fold for rich people to live, and i express me softly now. For all, or none. simple. More people should wake up.

So, 12 more cores for SETI! yay!
ID: 1895272 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 . . . 8 · Next

Message boards : Number crunching : Come on admit it, you have an E5-2670 crunching on Boinc and/or Seti@Home. How is it doing?


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.