Accéder au contenu.
Menu Sympa

starpu-devel - Re: [Starpu-devel] Starpu stats about disk, OOC, and swapping

Objet : Developers list for StarPU

Archives de la liste

Re: [Starpu-devel] Starpu stats about disk, OOC, and swapping


Chronologique Discussions 
  • From: Amani Alonazi <amani.alonazi@kaust.edu.sa>
  • To: Samuel Thibault <samuel.thibault@inria.fr>, Amani Alonazi <amani.alonazi@kaust.edu.sa>, "starpu-devel@lists.gforge.inria.fr" <starpu-devel@lists.gforge.inria.fr>, Hatem Ltaief <Hatem.Ltaief@kaust.edu.sa>
  • Subject: Re: [Starpu-devel] Starpu stats about disk, OOC, and swapping
  • Date: Mon, 14 Jan 2019 21:24:33 +0300
  • Authentication-results: mail3-smtp-sop.national.inria.fr; spf=None smtp.pra=amani.alonazi@kaust.edu.sa; spf=Pass smtp.mailfrom=amani.alonazi@kaust.edu.sa; spf=None smtp.helo=postmaster@mail-ed1-f47.google.com
  • Ironport-phdr: 9a23:Cu0+MBdPPXQqgsZc0ilUMx7VlGMj4u6mDksu8pMizoh2WeGdxcW9YR7h7PlgxGXEQZ/co6odzbaO4+a4ASQp2tWoiDg6aptCVhsI2409vjcLJ4q7M3D9N+PgdCcgHc5PBxdP9nC/NlVJSo6lPwWB6nK94iQPFRrhKAF7Ovr6GpLIj8Swyuu+54Dfbx9HiTahYr5+Ngm6oRnMvcQKnIVuLbo8xAHUqXVSYeRWwm1oJVOXnxni48q74YBu/SdNtf8/7sBMSar1cbg2QrxeFzQmLns65Nb3uhnZTAuA/WUTX2MLmRdVGQfF7RX6XpDssivms+d2xSeXMdHqQb0yRD+v9LlgRgP2hygbNj456GDXhdJ2jKJHuxKquhhzz5fJbI2JKPZye6XQds4YS2VcRMZcTyNOAo2+YIUPAeQPPvtWoZfhqFYVtxSyGROhCfnzxjNUhHL727Ax3eQ7EQHB2QwtB84DsHHOo9XzNacZTPy4zK3PzTXFdfxX2Db945XPfx87pPGDR7RwccvLxUY1EwPFiVOQppbkPjOUzekNsHKU7/d7VeKzkWIotRx+oiW1ysc3kIbEn4QYwU3H+yVh2Is5O8G0RUphbdOnEJZcrT+WO5d1T884TGxluiA3waAct5GhZigF0pEnygbfa/OZd4iI5QruVOOLLjd5gHJpYaywiAuv/US5xO3xWcu53ExFripCldnMuXQN2ALJ5sebTft9+1+t2TeJ1w/N9uFJOV44mbbfJpI7wbM9loAfvVrfEiL1gkn7g62bel0h+uey6uTnZrvmpoWbN49xkgzxKKEulda5AeQ5KAQBRWyb+eO61LL5/Ez5Qa5Hjv4snanEq5DaOd4XqbCkDA9IyIoj8AywAy+80NsEhXkHME5FeBWfgojyJ1HBOur4De+ig1i1jThk2e7GMaPlApjWKnjDkazhfapm60JGxwozw9df54hbCr4fOv7zW0nxtMbZDhAjKQC0zfznW51B0dYlUGeSD6vRDKPTu1aV4v5nd/KFYZIYvHDhK/kv7uPqkVc4n0UcdO+nx81TIFm9F/Jna2ydZnzlkNZJRW4Ptww/CuXuj1uLQz97ZH+vRL566z0mTpmvW8OLDJuxiaaZwWK3E4NbYkhCC0uQCjHneYKeVPpKaSSII8YnnCZOHey6V4Y7zQzruALkxr5PKuvP5jZeupzkztd4oezVjxA7szJuWYDV82iRQmQ8un8BRjI4lPRfqFBwzxGnzaF5jvJwFtVO/O8PXwsnc4XVmb9UEdf3DyvEd9GIAH+iRdOvEzB5Gts0ytkPJU95HNOrnBHr3CO3GKNTmrCWQoc9pPGPl0PtLtpwni6VnJIqiEMrF44WbTX/1/xPsjPLDouMqH230qOjdKASxinIrTvRzmOTp10eXQJtF77MDylGOhnm6O/h70aHdIeATKw9O1Idm82LN7BWLNDlkBNdTaW7YYmMUyeKg261QC2w6PaMYY7tIThP2SzcDA0Jn1lW8yrZcwc5ASilriTVCzk8TV8=
  • List-archive: <http://lists.gforge.inria.fr/pipermail/starpu-devel/>
  • List-id: "Developers list. For discussion of new features, code changes, etc." <starpu-devel.lists.gforge.inria.fr>

Thanks a lot, Samuel!

Hopefully last question :-) I'm using STARPU_DISK_SWAP_BACKEND=unistd. How can I analyze how much data transferred to disk during OOC computation? Shall I assume that anything exceeds the sustained BW that it runs at  the sustained BW?

I clearly notice the extra time required in executing each tasks with big/smaller STARPU_LIMIT_CPU_MEM. I just want to quantify the amount of transfer in/out to disk.

Many thanks!
A.

On Mon, Jan 14, 2019 at 8:53 PM Samuel Thibault <samuel.thibault@inria.fr> wrote:
Amani Alonazi, le lun. 14 janv. 2019 20:46:45 +0300, a ecrit:
> Indeed. But, StarPU statistics reports more than 90 MB/s, which is the
> sustained BW.

Ah, ok (please always specify what you think looks wrong :) ).

Yes, that may look a bit odd. But that can happen if you are not using
the unistd_o_direct backend, in which case the OS makes some caching.
Even with unistd_o_direct I wouldn't be surprised that the measurements
differ noticeably, since the Linux block scheduler also comes in etc.

Samuel


This message and its contents, including attachments are intended solely for the original recipient. If you are not the intended recipient or have received this message in error, please notify me immediately and delete this message from your computer system. Any unauthorized use or distribution is prohibited. Please consider the environment before printing this email.


Archives gérées par MHonArc 2.6.19+.

Haut de le page