<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
I was reasoning from the analogy that an incandescent bulb is less efficient than a LED bulb because more it generates more heat - more of the electricity goes into the infrared spectrum than the useful visible spectrum. Similar to the way that an electric
motor is more efficient than a combustion engine. <br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0);">
<br>
<hr tabindex="-1" style="display:inline-block; width:98%;">
<b>From:</b> Thomas Bellman<br>
<b>Sent:</b> Wednesday, December 18, 2019 7:47 PM<br>
<b>To:</b> Nanog@nanog.org<br>
<b>Cc:</b> Rod Beck<br>
<b>Subject:</b> Re: Energy Efficiency - Data Centers
<div><br>
</div>
</div>
<div class="BodyFragment"><font size="2"><span style="font-size:11pt;">
<div class="PlainText">On 2019-12-18 15:57, Rod Beck wrote:<br>
<br>
> This led me to wonder what is the inefficiency of these servers in data> centers. Every time I am in a data center I am impressed by how much> heat comes off these semiconductor chips. Looks to me may be 60% of the> electricity ends up as heat.<br>
What are you expecting the remaining 40% of the electricity ends up as?<br>
<br>
In reality, at least 99% of the electricity input to a datacenter ends up<br>
as heat within the DC. The remaining <1% turns into things like:<br>
<br>
- electricity and light leaving the DC in network cables (but will<br>
turn into heat in the cable and at the receiving end)<br>
- sound energy (noise) that escapes the DC building (but will turn<br>
into heat later on as the sound attenuates)<br>
- electric and magnetic potential energy in the form of stored bits<br>
on flash memory, hard disks and tapes (but that will turn into heat<br>
as you store new bits over the old bits)<br>
<br>
(I'm saying <1%, but I'm actually expecting it to be *much* less than<br>
one percent.)<br>
<br>
This is basic physics. First law of thermodynamics: you can't destroy<br>
(or create) energy, just convert it. Second law: all energy turns into<br>
heat energy in the end. :-)<br>
<br>
<br>
You are really asking the wrong question. Efficiency is not measured<br>
in how little of the input energy is turned into heat, but in how much<br>
*utility* you get out of a certain amount of input energy. In case of<br>
a datacenter, utility might be measured in number of database transac-<br>
tions performed, floating point operations executed, scientific articles<br>
published in Nature (by academic researchers using your HPC datacenter),<br>
or advertisments pushed to the users of your search engine.<br>
<br>
<br>
There is another efficiency number that many datacenters look at, which<br>
is PUE, Power Usage Effectiveness. That is a measure of the total energy<br>
used by the DC compared to the energy used for "IT load". The differece<br>
being in cooling/ventilation, UPS:es, lighting, and similar stuff.<br>
However, there are several deficiencies with this metric, for example:<br>
<br>
- IT load is just watts (or joules) pushed into your servers, and does<br>
not account for if you are using old, inefficient Cray 1 machines or<br>
modern AMD EPYC / Intel Skylake PCs.<br>
<br>
- Replace fans in servers with larger, more efficient fans in the rack<br>
doors, and the IT load decreases while the DC "losses" increase,<br>
leading to higher (worse) PUE, even though you might have lowered your<br>
total energy usage.<br>
<br>
- Get your cooling water as district cooling instead of running your own<br>
chillers, and you are no longer using electricity for the chillers,<br>
improving your PUE. There are still chillers run, using energy, but<br>
that energy does not show up on your DC's electricity bill...<br>
<br>
This doesn't mean that the PUE value is *entirely* worthless. It did<br>
help in putting efficiency into focus. There used to be datacenters<br>
that had PUE numbers close to, or even over, 2.0, due to having horribly<br>
inefficient cooling systems, UPS:es and so on. But once you get down<br>
to the 1.2-1.3 range or below, you really need to look at the details<br>
of *how* the DC achieved the PUE number; a single number doesn't capture<br>
the nuances.<br>
<br>
<br>
/Bellman<br>
<br>
</div>
</span></font></div>
</body>
</html>