Results 1 to 8 of 8

Thread: Linux Client Update

  1. #1

    Linux Client Update

    It's been a busy, but exciting, night at RieselSieve. We finally have a new Linux/x86 version that correctly reports its timing & status to the BOINC client. It also implements sr2sieve, rather than proth_sieve, so users should see a HUGE speedup.

    We also have introduced a Linux/ppc64 client. This will download from our server automatically, as long as you have compiled BOINC from source on your box. We plan to have a binary available soon. The new, from-source build is necessary because of the version string, powerpc64-unknown-linux-gnu, that it uses. Our server matches that & sends out a TRUE PPC sr2sieve version & wrapper. This also has the corrected linux progress & checkpointing capabilities. Thanks to BlisteringSheep for his help & access to a PPC box for us.

    As always, please visit the forums to notify us of bugs/changes.
    Thought you'd like to know, since the current linux clients are a bit shoddy.

    Bryan
    Bryan
    Stats Administrator
    BOINC Developer
    Riesel Sieve Project
    http://www.rieselsieve.com

  2. #2
    Join Date
    Apr 2005
    Location
    US
    Posts
    2,229
    Had some severe issues a few weeks ago on the boinc rieselsieve linux version. Still don't really know what happened that weekend when everything went nuts. Anyway I've detached from the project, and then deleted all the files/directories and did a fresh install using boinc_5.8.11 on some of my linux boxes. Will see how those go now before I put any more systems back on it.

    Are there any boinc linux users in here currently running this project?

  3. #3
    AMDave's Avatar
    AMDave is offline Seeker of the exit clause Moderator
    Site Admin
    Join Date
    Jun 2004
    Location
    Deep in a while loop
    Posts
    9,610
    Yes.
    I'm in for 2 linux clients atm (ver 5.4.x).
    There are returning wu's ok.

  4. #4
    Join Date
    Apr 2005
    Location
    US
    Posts
    2,229
    well so far things seem to be running okay. I reloaded more systems.

  5. #5
    Join Date
    Jul 2003
    Location
    Sydney, Australia
    Posts
    5,644
    No linux here; hard enough getting this thing to run under Windows XP

    I gave up on the LLR'ing as the tasks don't earn as many points as the Sieve'ing does. Got almost everything on BRS now and still only managing 4th in the dailies.

    Firing up the new e4300 now which replaces the XP2700+ that died last weekend:
    GA-965P-DQ6 m/b
    2GB Geil DDR2 Ultra 800 RAM
    Albatron GeForce 7950 GX2 1GB GPU
    CoolerMaster 430W p/s
    Antec SLK-3000B case
    Netgear WG111v2 USB wi-fi
    Seagate 320GB SATA 2 16MB cache HDD
    WinXP Pro SP2
    AVG 7.5 Pro


  6. #6
    Join Date
    Apr 2005
    Location
    US
    Posts
    2,229
    Looks like things are running back to normal for me now. I'm down one system that I've got to get back together that sorta croaked. Then I will be pulling out one of the older 1800+ systems and replacing it with an X2 3800. Thought about one of the C2D's, but then looking at the most bang for the buck, a couple X2's still outproduce a C2D on sieving. By the time the parts were assembled for a C2D, the cost was more than I cared to spend for one box.

    edit: Has onyone you know of done any benchmarking on this between the linux and windows clients? Of course they are both using different sieving clients, but I'm sort of curious as to what the actual comparative results are.
    Last edited by Brucifer; 03-04-2007 at 03:37 PM.

  7. #7
    AMDave's Avatar
    AMDave is offline Seeker of the exit clause Moderator
    Site Admin
    Join Date
    Jun 2004
    Location
    Deep in a while loop
    Posts
    9,610
    Brucifer

    the windows sieve is generally a jjsieve.exe client deployed via BOINC, whereas the linux sieve is still the old proth-sieve client, as you noted.

    The windows PIII and AMD CPU's saw an increase between 5 and 10 percent in keys per second when compared to the rate by proth-sieve on the same platform (Win+CPU). However the P4's continued to fare poorly. The recommendation was, as a result, for P4s to remain on LLR.

    Until that point, the output comparison between proth-sieve and windows was comparable on the PIIIs and AMDs, from my own experience, although I did find that the Semprons gave an excellent output per $$ in both windows and linux, although the linux platform slightly edged out in front by a couple of Kps (I think due to less overhead processes requesting CPU time).

    Some discovery on this was discussed here.
    http://www.rieselsieve.com/forum/viewtopic.php?t=971
    I believe there was more in-depth testing and benchmarking exposed in another thread, although it was likely covered by the lively discussion on the SOB forum.
    Last edited by AMDave; 03-05-2007 at 07:19 AM.

  8. #8
    Just to clarify - since the 5.45 version (5.30 for anything other than Linux/x86), we've been using the sr2sieve app which is a much closer competitor (I'd guess within 10%) of jjsieve.
    Bryan
    Stats Administrator
    BOINC Developer
    Riesel Sieve Project
    http://www.rieselsieve.com

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •