[E-impact] Re: Sustainable Procurement Guidelines for Data Centres and Servers from United Nations Environment Programme United for Efficiency (U4E)
Benjamin Schwarz <bs@ctoic.net> Fri, 05 September 2025 12:57 UTC
Return-Path: <bs@ctoic.net>
X-Original-To: e-impact@mail2.ietf.org
Delivered-To: e-impact@mail2.ietf.org
Received: from localhost (localhost [127.0.0.1]) by mail2.ietf.org (Postfix) with ESMTP id F0A635DFF39E for <e-impact@mail2.ietf.org>; Fri, 5 Sep 2025 05:57:23 -0700 (PDT)
X-Virus-Scanned: amavisd-new at ietf.org
X-Spam-Flag: NO
X-Spam-Score: -1.899
X-Spam-Level:
X-Spam-Status: No, score=-1.899 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001] autolearn=ham autolearn_force=no
Authentication-Results: mail2.ietf.org (amavisd-new); dkim=pass (2048-bit key) header.d=ctoic-net.20230601.gappssmtp.com
Received: from mail2.ietf.org ([166.84.6.31]) by localhost (mail2.ietf.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id si400GYWfoLD for <e-impact@mail2.ietf.org>; Fri, 5 Sep 2025 05:57:23 -0700 (PDT)
Received: from mail-wr1-x42b.google.com (mail-wr1-x42b.google.com [IPv6:2a00:1450:4864:20::42b]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature ECDSA (P-256) server-digest SHA256) (No client certificate requested) by mail2.ietf.org (Postfix) with ESMTPS id 014EE5DFF318 for <e-impact@ietf.org>; Fri, 5 Sep 2025 05:56:54 -0700 (PDT)
Received: by mail-wr1-x42b.google.com with SMTP id ffacd0b85a97d-3d1bf79d7acso1231561f8f.0 for <e-impact@ietf.org>; Fri, 05 Sep 2025 05:56:54 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ctoic-net.20230601.gappssmtp.com; s=20230601; t=1757077008; x=1757681808; darn=ietf.org; h=references:to:cc:in-reply-to:date:subject:mime-version:message-id :from:from:to:cc:subject:date:message-id:reply-to; bh=R9Xk4inlg8I/IWY1az3H+YLXgrKnhCXBnomJw4CFiiY=; b=DDdXL5aQ+cAsLVxNGAIKFx76yjV30EHs1UMvHkQ11usmNHkeHQXRDz38q1QLT4ToCv e8jgyPe24CyVya9HyPp+FcjRB7hKlU1BD+TqzfMTrkjtbfmP7phiiMk9XcS7CYK0Dv/G itNoeB+ufinj09d1xSdjnxrXKMiZ64RCfLjpM/KTc9lWevZgsKNJLYaupUceK1q2nB3b M0df4V7DLohOmY2bclzjEQHn6GT1ViBIyt4l4EhrccYEivECtKXQK43lPHidJPIXnL5E m139l/tubCOTSiIxYmij5g1lvFXYWGcsRcWKVpJ7gprq42oLu/01XRarMtB8umntrv6o Z+4Q==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1757077008; x=1757681808; h=references:to:cc:in-reply-to:date:subject:mime-version:message-id :from:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=R9Xk4inlg8I/IWY1az3H+YLXgrKnhCXBnomJw4CFiiY=; b=ZW6IZew1M0Totm3G0LlZ8MUTNDvvOnn3ZCClk/OoeuGNubNrKx96LnmUNDpFfsd6/K 1kptneJ/6NstCCbA93eBVQdDb2rR/GSPYpUMRgntLjAz5fRVZR4eIkttrS0TQeD0bdv0 /K6nK31M6/ZDRq3Dyr2maK4NsGC9aVv6fZInKuml9lp8lRg2UgzvKnfsso5ek/hXnCYZ MrhTWGaFMYBC8DmZ5wfEjrEgaV0eqwm+rZ764qlvTyuOfEf69kJtUFTjPSsbE9gC8pvl MG/Cc3MbQhl9bVvKWXaaX2bieRXNHBDBYLPatsTpPLqtdJSSIuK+HGakn/o0rlZGRbx0 +pJw==
X-Gm-Message-State: AOJu0YwJ8AVWOlWQWQ99Czzs60QxRDfvS9kVhx4RXTgGrgeHRCFP4gdG mnhMCLq/jzPe6n8xXBWtWOO45n1+FzP0P7y3eMLji/X+207w+PKUmWylCZmN1RtgE0nXSTEUR0Y r+TPw
X-Gm-Gg: ASbGncv49E9ggI6eNEl2O2QOx6xX2V74Qvj0XOAyo/EauixMOmUXB6fn5nCevEymuLc rzOAKlnEILdAd57T/xLWefhsigngFCpBuOa6HI6HCDcmundJ2UfPMOIzGpelvdWGvXb2mEVAAos PFAY84vmcxEdnLbkKTlypCFQlPFwMajkkTmcXCTT1Oh7HbjHEzCQpyQtUoxA2t59q4aDL4RmcY9 Lgx/XSbsD7dFNbPO6pirIF4ZsQlNZrRNLgpeF6dvUe/TDao+GltYRsBtUy/vD8PIHQoYYpzPUqQ wr2PJ3YqXn1a4R3wVRmZjNfHIU+ARufOAXxM91PsTd2Mjt5KuUjL1IYrKwNf0o4lmvgqTKWCrhr pU/a+7mZhKMDKtgW9J9CKchq0vFQY+vI8LfRhIgP/aFaFhbQPDnwJn92H
X-Google-Smtp-Source: AGHT+IEXkuYvY29xTgqB/1pFWppYkBcjw/JzJDSy1lAeIT30g4tMldkabHZ8nSFu5piozYo3pIsrNA==
X-Received: by 2002:a5d:5885:0:b0:3c7:84b4:e76 with SMTP id ffacd0b85a97d-3d1de4b90a8mr19753056f8f.25.1757077007860; Fri, 05 Sep 2025 05:56:47 -0700 (PDT)
Received: from smtpclient.apple (176-148-88-254.abo.bbox.fr. [176.148.88.254]) by smtp.gmail.com with ESMTPSA id 5b1f17b1804b1-45b7e7fec07sm319214385e9.10.2025.09.05.05.56.46 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Fri, 05 Sep 2025 05:56:47 -0700 (PDT)
From: Benjamin Schwarz <bs@ctoic.net>
Message-Id: <CA478E40-6C78-4582-AE3D-821BCB2F7D10@ctoic.net>
Content-Type: multipart/alternative; boundary="Apple-Mail=_4D3C2CB5-5764-493A-A9D3-88804BF4BAD9"
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3826.700.81\))
Date: Fri, 05 Sep 2025 14:56:36 +0200
In-Reply-To: <cd97d148-fa47-4d5f-a75f-bcc036e4418b@hefr.ch>
To: Sébastien Rumley <sebastien.rumley=40hefr.ch@dmarc.ietf.org>
References: <CAE4dcxmFewG0v1HTH7y3Vs9GqOkZAbcuZJv0UX=mQmYSGK_QSw@mail.gmail.com> <cd97d148-fa47-4d5f-a75f-bcc036e4418b@hefr.ch>
X-Mailer: Apple Mail (2.3826.700.81)
Message-ID-Hash: TQW57IUGUVYVN7QT4KDFWAO5Q7PAE4J7
X-Message-ID-Hash: TQW57IUGUVYVN7QT4KDFWAO5Q7PAE4J7
X-MailFrom: bs@ctoic.net
X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; emergency; loop; banned-address; member-moderation; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header
CC: e-impact@ietf.org
X-Mailman-Version: 3.3.9rc6
Precedence: list
Subject: [E-impact] Re: Sustainable Procurement Guidelines for Data Centres and Servers from United Nations Environment Programme United for Efficiency (U4E)
List-Id: Environmental impacts of the Internet <e-impact.ietf.org>
Archived-At: <https://mailarchive.ietf.org/arch/msg/e-impact/u9_WmkwlcKoMoHHAwPQNjLAg7wU>
List-Archive: <https://mailarchive.ietf.org/arch/browse/e-impact>
List-Help: <mailto:e-impact-request@ietf.org?subject=help>
List-Owner: <mailto:e-impact-owner@ietf.org>
List-Post: <mailto:e-impact@ietf.org>
List-Subscribe: <mailto:e-impact-join@ietf.org>
List-Unsubscribe: <mailto:e-impact-leave@ietf.org>
Hi Sebastien, Here are some recent data points that may help the discussion: Specialised vs. generic video encoding HW: In a recent blog <https://www.gl-systemhaus.de/en/blog/accelerated-video-encoding-how-specialised-hardware-can-triple-energy-efficiency>, I described that dedicated hardware was ~3× more energy-efficient at ≥66% system load and only ~2× at ~10% load, showing that once you reach the sweet spot (~70% load), we’re 50% more efficient. Operational target: A global CDN told us they aim for ~70% average server load for energy efficiency; I’m awaiting their dataset. Idle → peak deltas (processing only; peripherals constant, no PoE/USB/etc. changes): Entry-level switch: 70 W idle measured at the IETF Hackathon (Madrid) (+1 W per Gbps of traffic). Peak is (I believe) ~160 W (Hoping to redo these tests with Marisol, Artur and Luis on this list). Previous generation dedicated video-encoding server: 350 W idle → ~400 W peak (very low ratio). Generic PC encoding video: 100 W idle → 180 W Peak Raspberry PI: 3 W idle → 8 W peak. Home routers (5 to 25 W) (GoS tests and Orange White Paper): 0–10% variation between idle and fully loaded. Operator disclosure: DT publicly shared an overall network variation of ~10% in fixed and ~30% in mobile between idle and loaded. So, idle → peak varies from 0 (rare) to just over 2x in most things we’ve looked at; we need context to be more precise. Ben *** Benjamin Schwarz President, Greening of Streaming +33 674 787 954 bs@ctoic.net Book a meeting <https://tidycal.com/bs>  > On 5 Sep 2025, at 13:37, Sébastien Rumley <sebastien.rumley=40hefr.ch@dmarc.ietf.org> wrote: > > Hello Luis, > > Many thanks for this (with some delay), this is of high interest indeed. > > What I find particularly interesting is these metrics ITEEsv (IT Equipment Energy Efficiency for servers) and ITEUsv (IT Equipment Utilization for servers). They are IEC/ISO standards... but there is almost no mention of it in scientific literature. As someone in this group ever worked with this stuff? I guess this is pure operational stuff. Would it be too early to conclude that there is scant literature on data center operations ? > > I find interesting that the utilization rate is subject to objectives (>80% by 2031). I wonder if there is public data describing ITEUsv figures "in the wild". And I also wonder how such a metric is aggregated "data-center wide". I see how to calculate utilisation rate of a computer. But of mulitple ones... is that just an average ? But if there are many "small" servers with low utilisation and a few "big" servers high utilization, shouldn't we use a weighted average ? If anyone has some data on this, I am highly interested. > > Best, > > Sébastien > > > > On 30.07.2025 23:51, Luis M. Contreras wrote: >> Hi all, >> >> this could be a reference of interest. The United Nations, through the initiative called United for Efficiency (U4E) has recently released a white paper with focus on Sustainable Procurement Guidelines for Data Centres and Servers, which I think is of interest since a number of metrics and rations are described. The guidelines aim to represent a global best practice reference to support sustainable procurement and operation of data centres and servers. They balance technological feasibility, economic considerations, and environmental impact mitigation to foster a sustainable digital transformation aligned with international climate and development goals. >> >> The guidelines can be found here: 2025-U4E-Sustainable-Procurement-for-Data-Centers-and-Servers_Final.pdf <https://united4efficiency.org/wp-content/uploads/2025/02/2025-U4E-Sustainable-Procurement-for-Data-Centers-and-Servers_Final.pdf> >> >> In addition to that, slides from a webinar that presented that guidelines are available here: UNEP-U4E-Launch-Webinar-SPP-Guidelines-on-Data-Centres.pdf <https://united4efficiency.org/wp-content/uploads/2025/07/UNEP-U4E-Launch-Webinar-SPP-Guidelines-on-Data-Centres.pdf> >> >> Best regards >> >> Luis >> >> -- >> ___________________________________________ >> Luis M. Contreras >> contreras.ietf@gmail.com <mailto:contreras.ietf@gmail.com> >> luismiguel.contrerasmurillo@telefonica.com <mailto:luismiguel.contrerasmurillo@telefonica.com> >> Global CTIO unit / Telefonica >> >> > -- > Dr. Sébastien RUMLEY, Professor > iCoSys Institute, part of > HEIA-FR - School of Engineering and Architecture - Fribourg, part of > HES-SO - University of Applied Sciences and Arts Western Switzerland > -- > E-impact mailing list -- e-impact@ietf.org > To unsubscribe send an email to e-impact-leave@ietf.org
- [E-impact] Sustainable Procurement Guidelines for… Luis M. Contreras
- [E-impact] Re: Sustainable Procurement Guidelines… Sébastien Rumley
- [E-impact] Re: Sustainable Procurement Guidelines… Benjamin Schwarz
- [E-impact] Re: Sustainable Procurement Guidelines… Chris Adams
- [E-impact] Re: Sustainable Procurement Guidelines… Mark Butcher