Posts by MarkJ

21) Message boards : Number crunching : Do we have more data than volunteers? (Message 1871223)
Posted 5 Jun 2017 by Profile MarkJProject Donor
Post:
I'll take up some of the slack. Just brought 7000+ cores online on SETI@Home. The reason why is hilarious, but unfortunately I can't share it here.

Something like management want to stress test the disaster recovery site? Sounds like more cores than the Atlas cluster that Einstein use.

They aren't the latest kit but a year or two old judging by the Xeon v2 and v3's. Hate to think what the electricity bill would be like.

I need a shed like Wiggo, my loft gets too hot even in winter 😩
22) Message boards : Number crunching : CES 2017 -- AMD RYZEN CPU (Message 1871092)
Posted 4 Jun 2017 by Profile MarkJProject Donor
Post:
I've got a couple of Ryzen 1700's running such as this one running Linux. My current quandary is if I should replace the 1700's with 1700x or not. Apart from the initial cost of buying another CPU there is an extra 30 watts for the 1700x for about a 10 to 12% gain in productivity.

I did some experiments with the Einstein gravity wave tuning run on them. Running 16 at a time average run time was 33,000 to 37,000 seconds, when I limit it to 8 via an app_config they came in at 20,000 seconds. The app was their AVX and it got the Lo work units.

My current experiment is to see if leaving a single thread free improves performance or not. I'm running a single GPU with the Seti CUDA80 app so it can make use of the available thread but certainly doesn't need its own as it sleeps a fair bit (according to top).
23) Message boards : Number crunching : AMD's Upcoming Ryzen R9 "Threadripper" 16-core, 32-thread CPU (Message 1868334)
Posted 20 May 2017 by Profile MarkJProject Donor
Post:
For a no-cost solution reducing the deadline would help reduce the database bloat that needs frequent compression? While it may have an effect on the slower rigs they could run a smaller cache so they can meet the deadline.

Currently over at Einstein they're doing a Gravity wave tuning run with a 5 day deadline. Now that's obviously not practical for normal use but it certainly clears the tasks a lot quicker. Maybe 10 to 21 days is a better number to use.
24) Message boards : Number crunching : Help with Intel GPU on Arch Linuc (Message 1866694)
Posted 11 May 2017 by Profile MarkJProject Donor
Post:
I am running Debian. When I set up mine I had to get beignet 1.3.0 (beignet-opencl-icd) from Jessie backports to support the HD Graphics 530. I also needed to get boinc-client-opencl from there. You don't need mesa.

Given you're running Arch I am not sure how up to date their repos are.

As mentioned above running the iGPU will slow down the rest of the CPU tasks so its usually not worth the effort.
25) Message boards : Number crunching : You have to love early Xmas and B'day presents (even if you have to pay for them yourself). (Message 1864716)
Posted 30 Apr 2017 by Profile MarkJProject Donor
Post:
I've converted the farm to Linux. Up until Friday the two GTX1060's were running the Petri special and then I decommissioned the machines. I'm selling them off to get a couple of Ryzen's and the two GTX1060's will go in them.

In the mean time I have a couple of i3's running a GTX970 each with Petri special. I've ordered another pair of 1060's which will replace the 970's in order to cut down the electricity use a bit. Its nowhere near Christmas but that's my present to myself :-)
26) Message boards : Number crunching : Setting up a Linux machine to crunch CUDA80 for Windows users (Message 1861775)
Posted 16 Apr 2017 by Profile MarkJProject Donor
Post:
I got the cuda libraries that Petri posted earlier in this thread and put them in the /var/lib/boinc-client directory. They may not be needed. Make sure they're marked as executable and are owned by user boinc (do a chown boinc:boinc lib* command).

Apparently you don't need these to get it to recognise CUDA. You will of course need to put them and the app into the projects/setiathome folder along with an app_info.
27) Message boards : Number crunching : Setting up a Linux machine to crunch CUDA80 for Windows users (Message 1861619)
Posted 15 Apr 2017 by Profile MarkJProject Donor
Post:
Rather than using an outdated Ubuntu this is what I used with Debian. This will get you the latest Debian (Jessie), latest kernel (4.9) and the 7.6.33 BOINC client. This won't give you the CUDA80 app but you should be up and running with a CUDA capable machine after doing this.


Part 1 - Install Debian
I used the Debian 8.7 net install for this. You’ll need a thumb drive or a blank CD. Download it from http://www.debian.org/distrib/ and write the ISO image to CD or thumb drive.

Boot off the thumb drive or CD. It will start up the Debian installer

Install Debian select, SSH server and whatever desktop you prefer and remove all other selections. Once done it will reboot.


Part 2 - Install Nvidia software
Login as root, open a xterm window and type the following commands:

cd /etc/apt

nano sources.list (nano is a text editor)

Change “Jessie Main” lines to “jessie main contrib non-free” and add a Jessie-backports line. It should look like this when you're done. I'm using httpredir as it will pick the fastest server.

deb http://httpredir.debian.org/debian/ jessie main contrib non-free
deb http://security.debian.org/ jessie/updates main contrib non-free
deb http://httpredir.debian.org/debian/ jessie-updates main contrib non-free
deb http://httpredir.debian.org/debian/ jessie-backports main contrib non-free

Exit out of nano and save the file (Control-O followed by Control-X)

apt update

apt install –t jessie-backports firmware-realtek (if needed)

apt install –t jessie-backports linux-image-amd64

apt install –t jessie-backports nvida-kernel-dkms nvidia-smi nvidia-xconfig

apt install –t jessie-backports nvidia-opencl-icd (if you want OpenCL support)

nvidia-xconfig

sync

reboot


Part 3 – Install BOINC
login as root. Start xterm again and type the following commands:

apt install –t Jessie-backports boinc-nvidia-cuda boinc-manager

I got the cuda libraries that Petri posted earlier in this thread and put them in the /var/lib/boinc-client directory. They may not be needed. Make sure they're marked as executable and are owned by user boinc (do a chown boinc:boinc lib* command).

sync

reboot
28) Message boards : Number crunching : Water cooling the Breakthrough Listen CPUs and GPUs (Message 1861191)
Posted 13 Apr 2017 by Profile MarkJProject Donor
Post:
My GPU's are running on air. If I drool it will be a mixture of air and water :)

P.

Ahh a hybrid setup
29) Message boards : Number crunching : Looking for Linux x64 optimized apps (Message 1859180)
Posted 2 Apr 2017 by Profile MarkJProject Donor
Post:
An update on this. I have swapped all my iGPU machines over to Linux. While I have beignet installed I haven't defined an OpenCL app in my app_info as it slows the rest of the CPU tasks down. For the moment they're running as CPU only rigs. I have defined the astropulse app as well but AP work units are as rare as hens teeth :-)
30) Message boards : Number crunching : CES 2017 -- AMD RYZEN CPU (Message 1852763)
Posted 4 Mar 2017 by Profile MarkJProject Donor
Post:
I'm waiting delivery of a 1700, Asus x370 pro motherboard and 8GB RAM.

I think you may find that too little memory unless its only running Seti.

I had to upgrade the memory in all my i7-6700's to 16Gb to run the Einstein OM1CV work units, They can use up to 1.5Gb per work unit. Some of the VM based work units from other projects can also be a memory hog from what I hear, so I am ordering 32Gb with mine.
31) Message boards : Number crunching : CES 2017 -- AMD RYZEN CPU (Message 1852521)
Posted 3 Mar 2017 by Profile MarkJProject Donor
Post:
I have tried to order one from PC case gear except their credit card gateway determined I was behind a corporate firewall and kept blocking my payment. I was at work. After I rang them they said it blocks for 48 hours so I'm in back off until next week 😩
32) Message boards : Number crunching : Looking for Linux x64 optimized apps (Message 1852520)
Posted 3 Mar 2017 by Profile MarkJProject Donor
Post:
Looking at the AVX one its r3345 by Urs.

The stock app is r3584 so is more recent than the AVX app.


And how do they compare ?
Does someone have used the benchmark tool ?
Any hope to see an updated AVX package ?

Can't tell at the moment because when running stock about all I was getting were vlar work. Since then opt app has picked up most normal work so it's going to take a while. Also I was running OpenCL iGPU app with stock but don't have it in app_info at the moment so that might effect results.

Question for Raistmer or Urs: Does the stock Linux iGPU app have the accuracy options that the Windows version recently received or is it likely to produce invalids.
33) Message boards : Number crunching : Looking for Linux x64 optimized apps (Message 1852274)
Posted 2 Mar 2017 by Profile MarkJProject Donor
Post:
Looking at the AVX one its r3345 by Urs.

The stock app is r3584 so is more recent than the AVX app.
34) Message boards : Number crunching : Looking for Linux x64 optimized apps (Message 1852273)
Posted 2 Mar 2017 by Profile MarkJProject Donor
Post:
Ask for Petri software. It's rocket science !


:) Thanks.

His Linux machine does not have an NVIDIA GPU and my optimisations are not in the OpenCL version.

Petri

I do have a couple of GTX1060's under Windows. Looking forward to when the Petri special is ready for public consumption.
35) Message boards : Number crunching : Looking for Linux x64 optimized apps (Message 1852112)
Posted 1 Mar 2017 by Profile MarkJProject Donor
Post:
Yes I have looked and found the ones over at Lunatics, however they're dated Jan 2016. Looking at the AVX one its r3345 by Urs. The latest one on the list is a SoG beta app dated Feb 2017 (r3602). Is there anything more recent available for CPU or iGPU.

Machine is an i7 with Intel iGPU. It was a Windows machine 2 days ago. I'm trialling Linux on it. I finally got Beignet to recognise the iGPU and its done a bunch of multi-beam tasks using the stock apps. The iGPU is around the same speed as the prior Windows app but the CPU ones are somewhat slower.
36) Message boards : Number crunching : CES 2017 -- AMD RYZEN CPU (Message 1848634)
Posted 14 Feb 2017 by Profile MarkJProject Donor
Post:
I haven't been able to find anything that mentions how much memory the X370 chipset/CPU will support. I'm hoping (given the Ryzen 7 1700 has 8 cores/16 threads) that it will support up to 32Gb of DDR4.
37) Message boards : Number crunching : Linux (ARM processor) app and alternatives (Message 1848633)
Posted 14 Feb 2017 by Profile MarkJProject Donor
Post:
Is that fixed chirp version going to appear on beta as 8.06 then? Currently still getting 8.04 on beta as stock app on Pi2 and Pi3.
38) Message boards : Number crunching : CES 2017 -- AMD RYZEN CPU (Message 1848267)
Posted 12 Feb 2017 by Profile MarkJProject Donor
Post:
So that has me looking at a couple of Asus Prime X370 motherboard with Ryzen 7 1700 (8 core/12 thread 65w) machines to replace i7-4770K machines (which pull 140w just for the CPU).

Memory support is 2400Mhz which seems rather slow unless Asus give their overclocked memory options.
39) Message boards : Number crunching : Linux (ARM processor) app and alternatives (Message 1847488)
Posted 9 Feb 2017 by Profile MarkJProject Donor
Post:
Have attached a single Pi2 and a single Pi3 to Seti Beta to help out with testing. They've both picked up the 8.03 app and have four work units running.

Got blc05 vlar work on both Pi2 and Pi3.

Pi3 on 97% done after 23 hours and 6 mins so looks like it will be around 24 hours
Pi2 on 30% done after 23 hours and 8 mins so looks like 6 days.
40) Message boards : Number crunching : Linux (ARM processor) app and alternatives (Message 1847312)
Posted 8 Feb 2017 by Profile MarkJProject Donor
Post:
Have attached a single Pi2 and a single Pi3 to Seti Beta to help out with testing. They've both picked up the 8.03 app and have four work units running.


Previous 20 · Next 20


 
©2017 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.