From: "Andrei Berceanu berceanu@runbox.com" Received: from mx1.fz-rossendorf.de ([149.220.142.11] verified) by hzdr.de (CommuniGate Pro SMTP 6.2.7i) with ESMTP id 18893099 for picongpu-users@cg.hzdr.de; Mon, 03 Sep 2018 12:45:03 +0200 Received: from c1911.mx.srv.dfn.de (c1911.mx.srv.dfn.de [194.95.238.137]) by mx1.fz-rossendorf.de (Postfix) with ESMTP id 8539D3427A for ; Mon, 3 Sep 2018 12:46:00 +0200 (CEST) X-Virus-Scanned: Debian amavisd-new at mgw2-erl.srv.dfn.de Authentication-Results: mgw2-erl.srv.dfn.de (amavisd-new); dkim=pass (2048-bit key) header.d=runbox.com Received: from aibo.runbox.com (aibo.runbox.com [91.220.196.211]) by c1911.mx.srv.dfn.de (Postfix) with ESMTPS id E48674008C for ; Mon, 3 Sep 2018 12:45:58 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=runbox.com; s=rbselector1; h=Content-Type:To:Subject:Message-ID:Date:From:Reply-To: In-Reply-To:References:MIME-Version; bh=VsZAf7MC7tXRaG+Anggcd9uVh1bZizGF/6nvf3dpY8g=; b=kv0h8V3fhc+YbhC7gX9lmCsRPq x1xy9cEVfZFPOau8Z20qFN/dkAHOkCekZFLHn72/KRORP1yoFaGBXsF11a2nKg+OZLPG7qSZCbkSP a8NoCDLWKePptwoZbhPWdORikbNG8gt2BUVXZM1JlI9cIGUg7bESjUto8FLkOOSydb0pla3QHPDyi 9z+GQ2uyLqZJGiyX0a4DZfRbftbd/LLC0cJoVlMXTSIAAMRJi6NlbcXw0HxwVbqSTxBCVC+ObvY/1 AfCrMEpnTmDK2s+oi+t7ufPhYlp/7C3vg971mIkhNxeT5ARMq74KNAV/q36n5GJ03cwTae3YP8Cr6 xS+3yblQ==; Received: from [10.9.9.210] (helo=mailfront10.runbox.com) by mailtransmit02.runbox with esmtp (Exim 4.86_2) (envelope-from ) id 1fwmMh-0006Nh-Dj for picongpu-users@hzdr.de; Mon, 03 Sep 2018 12:45:51 +0200 Received: by mailfront10.runbox.com with esmtpsa (uid:811097 ) (TLS1.2:RSA_AES_128_CBC_SHA1:128) (Exim 4.82) id 1fwmMc-0004to-Qn for picongpu-users@hzdr.de; Mon, 03 Sep 2018 12:45:47 +0200 Received: by mail-qk1-f182.google.com with SMTP id f17-v6so69462qkh.4 for ; Mon, 03 Sep 2018 03:45:46 -0700 (PDT) X-Gm-Message-State: APzg51BQstZHzHlpSCwh4uTNBs3+WXYvGF2YiAEHu0UeTrLeHPucaHz6 8qsDFuzuZrTXjq1gl4f5PbN2SrYFahnpQu8FxAE= X-Google-Smtp-Source: ANB0VdabrKwR9pAuH+duBV5Mxa1dAqJec0M7LJW6tJ2Yq0eJgg89p1peW/znB5P/A4ayLtPfONKY+QuLmkoUiGYj5PA= X-Received: by 2002:ae9:eb58:: with SMTP id b85-v6mr24964126qkg.110.1535971545794; Mon, 03 Sep 2018 03:45:45 -0700 (PDT) MIME-Version: 1.0 References: In-Reply-To: Reply-To: berceanu@runbox.com Date: Mon, 3 Sep 2018 13:45:34 +0300 X-Gmail-Original-Message-ID: Message-ID: Subject: Re: [PIConGPU-Users] [PIConGPU-Users] PIConGPU performance across different GPUs To: picongpu-users@hzdr.de Content-Type: multipart/alternative; boundary="0000000000005f4fde0574f53e63" --0000000000005f4fde0574f53e63 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Thank you Ren=C3=A9. So are the 64 K80 boards on Hypnos 64 cards (with 2 GPUs each), or 64 GPUs (32 actual cards)? Best, Andrei On Mon, Sep 3, 2018 at 11:31 AM Ren=C3=A9 Widera r.widera@hzdr.de < picongpu-users@hzdr.de> wrote: > Hi Andrei, > > I have currently no performance data for the P100 available but if I > remember correct one P100 is two times faster than one k80 (with on K80 > I mean one card with two sockets). > For V100 I have currently no numbers in my head but due to improvements > of the shared memory atomics a V100 should be faster than a P100. > > best regards, > > Ren=C3=A9 > > On 08/31/2018 09:52 AM, Andrei Berceanu berceanu@runbox.com wrote: > > Dear all, > > > > In you experience, how does PIConGPU performance vary across different > > Tesla cards, such as the K80, P100 and V100? > > > > Is it fair to say, for instance, that one P100 is equivalent to two K80= s? > > > > Best regards, > > Andrei > > -- > Ren=C3=A9 Widera > Phone: +49 351 260 3543 > Institute of Radiation Physics > http://www.hzdr.de/crp > Helmholtz-Zentrum Dresden - Rossendorf (HZDR) > Bautzner Landstr. 400 | 01328 Dresden | Germany > Board of Directors: Prof. Dr. Dr. h. c. Roland Sauerbrey, Dr. Ulrich Breu= er > Company Registration Number VR 1693, Amtsgericht Dresden > > ############################################################# > This message is sent to you because you are subscribed to > the mailing list . > To unsubscribe, E-mail to: > To switch to the DIGEST mode, E-mail to > To switch to the INDEX mode, E-mail to > Send administrative queries to > > > --0000000000005f4fde0574f53e63 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Thank you Ren=C3=A9.
So are the 64 K80 boards on= Hypnos 64 cards (with 2 GPUs each), or 64 GPUs (32 actual cards)?

Best,
Andre= i

On Mon, Sep 3, 2018 at 11:31 AM Ren= =C3=A9 Widera r.widera@hzdr.de <= picongpu-users@hzdr.de> wr= ote:
Hi Andrei,

I have currently no performance data for the P100 available but if I
remember correct one P100 is two times faster than one k80 (with on K80 I mean one card with two sockets).
For V100 I have currently no numbers in my head but due to improvements of the shared memory atomics a V100 should be faster than a P100.

best regards,

Ren=C3=A9

On 08/31/2018 09:52 AM, Andrei Berceanu berceanu@runbox.com wrote:
> Dear all,
>
> In you experience, how does PIConGPU performance vary across different=
> Tesla cards, such as the K80, P100 and V100?
>
> Is it fair to say, for instance, that one P100 is equivalent to two K8= 0s?
>
> Best regards,
> Andrei

--
Ren=C3=A9 Widera
Phone: +49 351 260 3543
Institute of Radiation Physics
htt= p://www.hzdr.de/crp
Helmholtz-Zentrum Dresden - Rossendorf (HZDR)
Bautzner Landstr. 400 | 01328 Dresden | Germany
Board of Directors: Prof. Dr. Dr. h. c. Roland Sauerbrey, Dr. Ulrich Breuer=
Company Registration Number VR 1693, Amtsgericht Dresden

#############################################################
This message is sent to you because you are subscribed to
=C2=A0 the mailing list <picongpu-users@hzdr.de>.
To unsubscribe, E-mail to: <picongpu-users-off@hzdr.de>
To switch to the DIGEST mode, E-mail to <picongpu-users-digest@hzdr.de> To switch to the INDEX mode, E-mail to <picongpu-users-index@hzdr.de>
Send administrative queries to=C2=A0 <picongpu-users-request@hzdr.de>

--0000000000005f4fde0574f53e63--