From mboxrd@z Thu Jan 1 00:00:00 1970 From: ludo@gnu.org (Ludovic =?utf-8?Q?Court=C3=A8s?=) Subject: Re: Guix on clusters and in HPC Date: Thu, 20 Oct 2016 16:17:09 +0200 Message-ID: <87lgxj9kbe.fsf@gnu.org> References: <87r37divr8.fsf@gnu.org> <20161019.091750.442073013177424288.post@thomasdanckaert.be> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:40639) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bxE9j-0004q3-CS for guix-devel@gnu.org; Thu, 20 Oct 2016 10:17:16 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bxE9f-0007VL-EW for guix-devel@gnu.org; Thu, 20 Oct 2016 10:17:15 -0400 In-Reply-To: <20161019.091750.442073013177424288.post@thomasdanckaert.be> (Thomas Danckaert's message of "Wed, 19 Oct 2016 09:17:50 +0200 (CEST)") List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: "Guix-devel" To: Thomas Danckaert Cc: guix-devel@gnu.org Thomas Danckaert skribis: > From: ludo@gnu.org (Ludovic Court=C3=A8s) > Subject: Guix on clusters and in HPC > Date: Tue, 18 Oct 2016 16:20:43 +0200 > >> So I=E2=80=99ve come up with an initial list of work items going from the >> immediate needs to crazy ideas (batch scheduler integration!) that >> hopefully make sense to cluster/HPC people. I=E2=80=99d be happy to get >> feedback, suggestions, etc. from whoever is interested! > > Here's a semi-crazy suggestion: some HPC people really like the intel > compiler/math kernel library, so a way to switch toolchains on > packages, and somehow integrate =E2=80=9Cforeign=E2=80=9D toolchains woul= d suit > them. But that might be =E2=80=9Cadvocating the use of non-free software= =E2=80=9D (in > fact I almost feel like a heretic for bringing this up :-) ...). Yeah, I=E2=80=99m aware of this. In section 5 of , we wrote: GNU Guix does not provide proprietary software packages. Unfortunately, proprietary software is still relatively common in HPC, be it linear algebra libraries or GPU support. Yet, we see it as a strength more than a limitation. Often, these =E2=80=9Cblack boxes=E2=80= =9D inherently limit reproducibility=E2=80=94how is one going to reproduce a software environment without permission to run the software in the first place? What if the software depends on the ability to =E2=80=9Ccall home=E2=80=9D to function at all? More importantly, we view reproducible software environments and reproducible science as a tool towards improved and shared knowledge; developers who deny the freedom to study and modify their code work against this goal. As heretic as it may seem in HPC circles ;-), I stand by this. Regarding GPU support, there=E2=80=99s light on the horizon with GCC=E2=80= =99s new PTX backend, OpenACC support, and with libgomp=E2=80=99s offloading support. We=E2=80=99ll see! Ludo=E2=80=99.