Posts by ivan


log in
1) Message boards : Number crunching : Huge 8000+ hour work units (Message 1669575)
Posted 20 hours ago by Profile ivan
Wasn't a computer back in 1975 called ... a calculator? LOL

Not necessarily. This is what I was using back in 1975.
2) Message boards : Number crunching : Intel® iGPU AP bench test run (e.g. @ J1900) (Message 1668132)
Posted 4 days ago by Profile ivan
I've noticed lately that some GPU tasks on my Win10 J1900 are stalling at some point (e.g. the first one I noted was at a suspicious 66.677%) while the CPU total keeps marching on. If I restart BOINC then the Progress counter drops back a little but the CPU total drops back a lot (presumably to the time the Progress stalled) and again the Progress ticks up until it hits the previous stall point and stops there as CPU keeps building.
Has anyone else seen that? Or know a cure? I'm just running bare Lunatics with no tweaks.
3) Message boards : Number crunching : Problem with the SSL CA cert (Message 1665457)
Posted 10 days ago by Profile ivan
There was a problem at GPUGrid recently with an outdated ca-bundle.crt file - it's supplied by BOINC, and installed in the BOINC program directory. The ca-bundle.crt from a newer BOINC can be dropped in as a direct replacement

See: https://www.gpugrid.net/forum_thread.php?id=3846&nowrap=true#40528
(Crunch3r was wrong - you don't even have to re-start BOINC)

Thanks, Richard; I had two Linux machines at work that were suffering the problem and the download fixed them. There's another Win10 machine in my office not reported in yet tonight, but I can't log-on to it from here to check why.

OK, it's reported in on its own, so all my machines are current after the maintenance break.
4) Message boards : Number crunching : Problem with the SSL CA cert (Message 1665439)
Posted 10 days ago by Profile ivan
There was a problem at GPUGrid recently with an outdated ca-bundle.crt file - it's supplied by BOINC, and installed in the BOINC program directory. The ca-bundle.crt from a newer BOINC can be dropped in as a direct replacement

See: https://www.gpugrid.net/forum_thread.php?id=3846&nowrap=true#40528
(Crunch3r was wrong - you don't even have to re-start BOINC)

Thanks, Richard; I had two Linux machines at work that were suffering the problem and the download fixed them. There's another Win10 machine in my office not reported in yet tonight, but I can't log-on to it from here to check why.
5) Message boards : Number crunching : Best sub $200 video card for SETI? (Message 1665183)
Posted 11 days ago by Profile ivan
As an aside, I have a couple of 750Ti running great guns and to be honest, my research threw up that not only was the number crunching ability of the 750Ti pretty damned good, the power requirements are virtually nothing compared to some other cards.

They're PCIe bus powered, that could well be a biggie for people with smaller PSUs

Not only there; I was given a [(hp)] "Enterprise" PC at work. It's proprietary-everything, including the power supply, so no extra PCIe power connectors. I had to wait until the 750Ti came out to replace the standard (IIRC) 440. :-(
6) Message boards : Number crunching : Panic Mode On (97) Server Problems? (Message 1665177)
Posted 11 days ago by Profile ivan
We have yet to find it, but we are looking.

A bit of documentary.

Dancing in the Dark...

Might keep you interested whilst the project is in intermission.
Very fine documentary on what we are doing.

Meow!

Yes, a very good documentary, but it gave me a few problems. I haven't cut my hair for two years, I hadn't shaved since last Easter. In short, I looked a lot like John Ellis in that doco. Everyone who knew I worked at CMS assumed it was in fact me in the prog. I shaved my beard a couple of weeks earlier than I normally do. Terrible to be mistaken for a theoretical physicist...
7) Message boards : Number crunching : whats up with the badges (Message 1662586)
Posted 17 days ago by Profile ivan
Yeah, and I think I crossed into 1% territory quite a while ago, if indeed it's based on RAC. I think I'm in the top 300's somewhere ...

If you look at your stats page you'll see all types of badge on one page, yet the hover-over shows them as "Top n% in average credit" even though the display is sorted by average credit. Something's borked.
8) Message boards : Number crunching : WU In Progress Limit? (Message 1662389)
Posted 17 days ago by Profile ivan
100 work units for your CPU and 100 for each GPU that you have. Total for your machines is 200.

If you had 2 GPUs and the CPU then it would be 300..

3 GPUs and cpu then 400....

You get the idea.

Which is, in a way, annoying... I have two machines with 20 cores each (and I've seen some similar with HyperThreading(TM) turned on so they run 40 WUs at once) and they are by no means the biggest CPU-count machines around. If I could work out how to get S@H running on my Xeon Phi co-processor... It has 60 cores and 240 threads, so it could never be filled under the current regime (but realistically it has only 4-5 GB of memory for all those threads to share). The next-generation Xeon Phi will actually have the processor as the onboard CPU rather than a co-processor -- which makes it look a much better prospect for BOINC/S@H. I hope I'm not breaking any NDAs when I say that it will have between 60 and 72 cores, again each with four hardware threads, but this time it will have access to system memory (as well as 16 GB of fast on-board RAM) so running that many jobs will be feasible.
9) Message boards : Number crunching : Windows 10 may put SETI@home in Danger (Message 1655378)
Posted 21 Mar 2015 by Profile ivan
Miss the screensaver myself...

I do too.

Provided it blanks after 5 minutes to avoid wasting CPU cycles.

Indeed. I had a cow-orker once who used the "random" screensaver in RHEL4. He was often out of the room, screensaver running but (LCD) screen blank(!). From time to time I'd hear his CPU fan spin up and kick into high gear, and if I went over and wiggled his mouse one particular screensaver would be running -- Pipes IIRC.
What a waste of CPU cycles!
10) Message boards : Number crunching : Nvidia Titan X (Message 1655376)
Posted 21 Mar 2015 by Profile ivan
X it is then. Awesome, it's cheap!

(relatively speaking and for what it is...)

Cheap? In the US maybe ($999). Over here it's coming in around £900 ($US1,350).
http://www.ebuyer.com/store/Components/cat/Graphics-Cards-Nvidia/subcat/NVIDIA-GTX-TITAN
11) Message boards : Number crunching : Some sad news I need to pass on... (Message 1655372)
Posted 21 Mar 2015 by Profile ivan
Condolences to you and your family, Richard. Everything else has been said, probably better than I could myself.
12) Message boards : Number crunching : Banana Pi (Message 1653249)
Posted 15 Mar 2015 by Profile ivan
My Jetson build seems to be running OK!
http://setiathome.berkeley.edu/results.php?hostid=7354060&offset=0&show_names=0&state=4&appid=
13) Message boards : Number crunching : CPU Power Usage Varies Significantly With RAM Frequency (Message 1652837)
Posted 14 Mar 2015 by Profile ivan
To program a GPU efficiently one must do just that right now in his/hers mind.

Yes, GPU programming is a whole new mindset -- two talented programmers before me failed to get our analysis software running on a FPS vector-processor (TRIUMF, 1984) before I tried and succeeded. Looking today at the theoretical and S@H-reported performance of my Tegra GPU -- 327 GFLOPS and 14.5 GFLOPS respectively (double the S@H figure as I run two jobs at once) -- it does appear that there is some way to go in extracting ultimate performance. In my experience, this may need new algorithms.
Mind you, the same is true of complex CPU software, as this 2007 report on my current project's software shows -- I doubt programmers or compilers have gotten significantly more proficient in the interim.
14) Message boards : Number crunching : Banana Pi (Message 1652829)
Posted 14 Mar 2015 by Profile ivan
BOINC doesn't seem to be reliably able to determine the number of CPU cores -- sometimes it thinks 1, sometimes 2, finally settled on 4. I've never worked out how to use the fifth core in Ubuntu ...

You have to be able to use <ncpus>N</ncpus> in cc_config.xml to force/tell BOINC you have N CPUs (cores)

Yeah, thanks, I guess that'll solve it!
<test>
14-Mar-2015 12:03:03 [---] CUDA: NVIDIA GPU 0: GK20A (driver version unknown, CUDA version 6.0, compute capability 3.2, 1746MB, 808MB available, 327 GFLOPS peak) 14-Mar-2015 12:03:03 [SETI@home] Found app_info.xml; using anonymous platform 14-Mar-2015 12:03:03 [---] Host name: tegra-ubuntu 14-Mar-2015 12:03:03 [---] Processor: 4 ARM ARMv7 Processor rev 3 (v7l) 14-Mar-2015 12:03:03 [---] Processor features: swp half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt 14-Mar-2015 12:03:03 [---] OS: Linux: 3.10.24-g6a2d13a 14-Mar-2015 12:03:03 [---] Memory: 1.71 GB physical, 0 bytes virtual 14-Mar-2015 12:03:03 [---] Disk: 58.55 GB total, 53.88 GB free 14-Mar-2015 12:03:03 [---] Local time is UTC +0 hours 14-Mar-2015 12:03:03 [---] Config: simulate 4 CPUs 14-Mar-2015 12:03:03 [---] Config: use all coprocessors

top - 12:06:11 up 4 days, 1:54, 5 users, load average: 5.15, 4.68, 4.86 Tasks: 209 total, 6 running, 202 sleeping, 0 stopped, 1 zombie %Cpu(s): 0.2 us, 10.4 sy, 89.1 ni, 0.2 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st KiB Mem: 1788136 total, 1697140 used, 90996 free, 47592 buffers KiB Swap: 0 total, 0 used, 0 free. 123656 cached Mem PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 10604 ubuntu 39 19 73156 70636 1780 R 93.4 4.0 2:33.12 setiathome-7.28 10605 ubuntu 39 19 73300 70868 1772 R 89.4 4.0 2:31.85 setiathome-7.28 10607 ubuntu 39 19 70684 68452 1776 R 73.7 3.8 2:32.95 setiathome-7.28 10606 ubuntu 39 19 73220 70700 1788 R 72.3 4.0 2:31.89 setiathome-7.28 10608 ubuntu 30 10 403984 99964 17072 S 25.3 5.6 0:49.50 setiathome_x41z 10609 ubuntu 30 10 406208 102680 17376 R 24.7 5.7 0:48.96 setiathome_x41z

Sofa so good...
</test>
Cheers!
15) Message boards : Technical News : Results (Mar 11 2015) (Message 1652671)
Posted 14 Mar 2015 by Profile ivan
Oh now it looks like georgem is crashing too with the same NFS errors. Ha ha. Happy Friday afternoon...

- Matt

Ouch, everything bad happens on Friday near quitting time. :(

In my extensive experience as an experimental physicist at various institutes around the world, everything bad happens on a Friday afternoon after the technicians have left, but before the scientists start knocking off for the weekend.
16) Message boards : Number crunching : Banana Pi (Message 1652549)
Posted 13 Mar 2015 by Profile ivan
One Note, If when compiling the app, If it fails with an error, check the client sub-directory anyway for the compiled app, there are two Bugs with compiling seti_boinc from head,

If you use a more minimal configure, it fails with the following, but the app is still compiled:

In file included from <command-line>:0:0:
//home/stephen/boinc/lib/../config.h:732:0: note: this is the location of the previous definition
#define VERSION "7.5.0"
^
readwu.cpp:47:20: fatal error: export.h: No such file or directory
#include "export.h"

^
compilation terminated.
make[2]: *** [readwu_so-readwu.o] Error 1
make[2]: Leaving directory `/home/stephen/seti_boinc/tools'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/stephen/seti_boinc'
make: *** [all] Error 2


Claggy

You're right, the executable was there. Running now, tho' BOINC doesn't seem to be reliably able to determine the number of CPU cores -- sometimes it thinks 1, sometimes 2, finally settled on 4. I've never worked out how to use the fifth core in Ubuntu, but since it's less powerful and the "real" CPUs look like they'll take about 100,000 seconds per WU I doubt that it would bring me much anyhow.
17) Message boards : Number crunching : Banana Pi (Message 1651809)
Posted 11 Mar 2015 by Profile ivan
Thanks for your answer!
In the meantime I installed the most Boinc packages, include boinc-dev. I think this is what you mean. Nothing happend, still "boinc not found".

You need to tell the ./configure where your BOINC headers are located. E.g.
[tegra-ubuntu:seti_boinc] > head -50 config.log This file contains any messages produced by compilers while running configure, to aid debugging if configure makes a mistake. It was created by setiathome_v7 configure 7.28, which was generated by GNU Autoconf 2.69. Invocation command line was $ ./configure BOINCDIR=../boinc --enable-fast-math --disable-graphics ## --------- ## ## Platform. ## ## --------- ## hostname = tegra-ubuntu uname -m = armv7l uname -r = 3.10.24-g6a2d13a uname -s = Linux uname -v = #1 SMP PREEMPT Fri Apr 18 15:56:45 PDT 2014
18) Message boards : Number crunching : thinking about upgrading GPU's (Message 1651807)
Posted 11 Mar 2015 by Profile ivan
Okay i have been running GTX 570 in SLI for a few years now. I was thinking about moving up to the GTx 980 but one at time (might do SLI again later in the year).

so the Q: is the 980 that much better than the old 570's i have now.

http://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units:
SP GFlOpS Power (W) GFlOpS/W GTX 570 1405.4 219 6.41 GTX 980 4612 165 28.0
19) Message boards : Number crunching : Win 10 will be a FREE Upgrade to Win 7 & 8 Users!!! (Message 1651129)
Posted 9 Mar 2015 by Profile ivan
Oops, I forgot, is anyone crunching on Windows 10 yet?

Yes. :-)
20) Message boards : Number crunching : NVIDIA SHIELD! I can't wait to run SETI on it! (Message 1649410)
Posted 5 Mar 2015 by Profile ivan
Multibeam support (at least) for that could be relatively quick, since it should I guess run Cuda natively (taking the specs at face value). Should manage OpenCL too, though I haven't read the Tegra SDK details on that. For My Nexus 9, I'm having to look at the viability of a Renderscript port as well, because Google decided to go with their own language instead of using Cuda and/or OpenCL.

Guess it's time to ask (again) if anyone's managed to get MB CPU tasks running on a Tegra TK1 Jetson. I got the CUDA version running relatively easily, and my digital hologram reconstruction, but when I tried the CPU version -- it runs fine manually, but crashes when run as a BOINC task. I should really look at it again but at the moment I'm up in my all in assigators trying to bridge the gap between xxxx IT and yyy scientists to get yyy@Home running for volunteer computing. (Of course, most of you know what xxxx and yyy are, but I'd better be circumspect.) With a little bit of luck, we'll have things ready for the public in late autumn; with a lot of luck, earlier!


Next 20

Copyright © 2015 University of California