<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);">
Well, the fact that a data center generates a lot of means it is consuming a lot of electricity. It is probably a major operating expense. And while it may be efficient given current technology standards, it naturally leads to the question of how we can do
better. <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);">
Thermodynamics is only part of the picture. The other part is economics. If you see a lot of heat being produced and it is not the intended output, then it is anatural focus for improvement. My guess is that a lot of corporate research is going into trying
to reduce chip electricity consumption. <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);">
So my gut feeling might still be relevant. It is about the level of energy consumption, not just the fact electricity becomes disorderly molecular gyrations.
<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 9:57 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 20:06 CET, Rod Beck wrote:<br>
<br>
> I was reasoning from the analogy that an incandescent bulb is less<br>
> efficient than a LED bulb because more it generates more heat - more<br>
> of the electricity goes into the infrared spectrum than the useful<br>
> visible spectrum. Similar to the way that an electric motor is more<br>
> efficient than a combustion engine.<br>
<br>
Still, you should not look at how much heat you get, but how much<br>
utility you get. Which for a lighting source would be measured in<br>
lumens within the visible spectrum.<br>
<br>
If you put in 300 watt of electricity into a computer server, you<br>
will get somewhere between 290 and 299 watts of heat from the server<br>
itself. The second largest power output will be the kinetic energy<br>
of the air the fans in the server pushes; I'm guestimating that to<br>
be somewhere between 1 and 10 watts (and thus my uncertainty of the<br>
direct heat output above). Then you get maybe 0.1 watts of sound<br>
energy (noise) and other vibrations in the rack. And finally, less<br>
than 0.01 watts of light in the network fibers from the server<br>
(assuming dual 40G or dual 100G network connections, i.e. 8 lasers).<br>
<br>
Every microwatt of electricity put into the server in order to toggle<br>
bits, keeping bits at their current value, transporting bits within<br>
and between CPU, RAM, motherboard, disks, and so on, will turn into<br>
heat *before* leaving the server. The only exception being the light<br>
put into the network fibers, and that will be less than 10 milliwatts<br>
for a server.<br>
<br>
All inefficiencies in power supplies, power regulators, fans, and<br>
other stuff in the server, will become heat, within the server.<br>
<br>
So your estimate of 60% heat, i.e. 40% *non*-heat, is off by at<br>
least a factor ten. And the majority of the kinetic energy of<br>
the air pushed by the server will have turned into heat after just<br>
a few meters...<br>
<br>
So, if you look at how much heat is given off by a server compared<br>
to how much power is put into it, then it is 99.99% inefficient. :-)<br>
<br>
But that's just the wrong way to look at it.<br>
<br>
In a lighting source, you can measure the amount of visible light<br>
given off in watts. In an engine (electrical, combustion or other-<br>
wise), you can measure the amount of output in watts. So in those<br>
cases, efficiency can be measured in percent, as the input and the<br>
output are measured in the same units (watts).<br>
<br>
But often a light source is better measured in lumens, not watts.<br>
Sometimes, the torque, measured in Newton-meters, is more relevant<br>
for an engine. Or thrust, measured in Newtons, for a rocket engine.<br>
Then, dividing the input (W) with the output (lm, Nm, N) does not<br>
give a percentage.<br>
<br>
Similarly, the relevant output of a computer is not measured in<br>
watts, but in FLOPS, database transactions/second, or web pages<br>
served per hour.<br>
<br>
Basically, the only time the amount of heat given off by a computer<br>
is relevant, is when you are designing and dimensioning the cooling<br>
system. And then the answer is always "exactly as much as the power<br>
you put *into* the computer". :-)<br>
<br>
<br>
/Bellman<br>
<br>
</div>
</span></font></div>
</body>
</html>