Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • May 02 2020 14:21
    ernstbrunr51 commented #278
  • Apr 29 2020 20:49
    yurivict opened #278
  • Apr 24 2020 18:55
    ernstbrunr51 closed #277
  • Apr 24 2020 18:55
    ernstbrunr51 commented #277
  • Apr 24 2020 18:04
    poulson commented #277
  • Apr 24 2020 10:34
    ernstbrunr51 opened #277
  • May 14 2019 01:33
    poulson closed #276
  • May 14 2019 01:33
    poulson commented #276
  • May 14 2019 01:30

    poulson on master

    Update README.md (compare)

  • May 14 2019 01:24

    poulson on master

    added logo (compare)

  • May 10 2019 21:37
    jedbrown commented #276
  • May 10 2019 21:25
    poulson commented #276
  • May 10 2019 21:21
    jedbrown commented #276
  • May 10 2019 17:08
    poulson commented #276
  • May 10 2019 17:03
    tesch1 opened #276
  • Apr 10 2019 13:18
    JM1 commented #275
  • Apr 10 2019 13:16
    JM1 synchronize #275
  • Apr 10 2019 08:56
    JM1 opened #275
  • Mar 06 2019 03:47
    Raviteja1996 closed #274
  • Mar 05 2019 05:46
    Raviteja1996 opened #274
Jack Poulson
@poulson
I doubt that the hang is Elemental-specific
it might be worth checking that the ABI's all matched; hopefully there wasn't a mix-up between 64-bit int and 32-bit int versions
Ryan H. Lewis
@rhl-
@poulson: how would you diagnose that?
@poulson: I think I see the problem, I think you are right, it looks like the code is reading the matrix in wrong.
reading in row major..
Ryan H. Lewis
@rhl-
@poulson: can you merge my PR?
Jack Poulson
@poulson
Let me take a quick look
Also:
https://github.com/Homebrew/homebrew-science/pull/4118#issuecomment-242756079
I had added line notes to your PR
They seem to not be addressed yet
Ryan H. Lewis
@rhl-
Maybe I misunderstood, I thought I did.
@poulson: see, releases are necesary. the world runs on them. time to do it.
Jack Poulson
@poulson
Need to run D&C EVD and DVD at scale at least once before releasing
Colleague at Julich claims something is not working in their driver since I made the switch, but it is a significant possibility that the problem is on their end
Ryan H. Lewis
@rhl-
sounds like we have a release candidate.
Ryan H. Lewis
@rhl-
@poulson: turned out the problem was due to File I/O, everything is working as expected now.
Ryan H. Lewis
@rhl-
@poulson: ping?
Jack Poulson
@poulson
Pong
I would like to get a Henry/Walker/Dongarra style distributed Hessenberg QR impl' in before the official release so that dropping Scalapack as a default dependency does not sacrifice any serious functionality
This will help with the Debian package
I think I can have it working by tomorrow night if I scramble
Ryan H. Lewis
@rhl-
Is that a bulge chasing type thing?
Jack Poulson
@poulson
yes
Ryan H. Lewis
@rhl-
I actually wrote some code for that once
@poulson: lets set a date and release version number and then lets create a milestone in GitHub
Ryan H. Lewis
@rhl-
@poulson: found the code: https://github.com/rhl-/qr
this is not necessarily efficient
Ryan H. Lewis
@rhl-
I think its only a distributed hessenberg too
i think we never finished the actual bulge chasing
Jack Poulson
@poulson
do you mean it was a distributed reduction to Hessenberg form?
Ryan H. Lewis
@rhl-
yeah
Jack Poulson
@poulson
what kind of data distribution?
Ryan H. Lewis
@rhl-
each processor gets a block
its not much code.. and i'm sure you could improve upon it greatly
its been a long time since I thought about that
this was for something at Stanford
lexing's class or something
its not really high quality
you presumably have a distributed hessenberg already
Jack Poulson
@poulson
even if it isn't, it's an impressive goal
yes, Elemental has distributed blocked algorithms for reducing to Hessenberg form
Ryan H. Lewis
@rhl-
this used boost::ublas
i didn't know about elemental then
Jack Poulson
@poulson
there are sequential Aggressive Early Deflation Hessenberg QR algorithms but not yet a distributed version
the right paper to start on is http://www.netlib.org/utk/people/JackDongarra/PAPERS/140_2002_a-parallel-implementation-of-the-nonsymmetric-qr-algorithm.pdf
Ryan H. Lewis
@rhl-
yeah so the QR iteration looks like its only serial, and the hessenberg is distributed
Ryan H. Lewis
@rhl-
So, I wanted to take a stab at the documentation
i was wondering if you have seen/considered cldoc ?
Jack Poulson
@poulson
does it support LaTeX?