gwax (Posts about computers)
https://gwax.com/
en
Contents ©20052024 <a href="link://slug/aboutperson">George LeslieWaksman</a>
<a rel="license" href="https://creativecommons.org/licenses/bysa/4.0/">
<img
src="https://i.creativecommons.org/l/bysa/4.0/80x15.png"
alt="Creative Commons License BYSA"
style="border:0;margin:0;"
/>
</a>
Sat, 20 Jan 2024 00:39:04 GMT
Nikola (getnikola.com)
http://blogs.law.harvard.edu/tech/rss

On airplane bandwidth and latency
https://gwax.com/blog/2009/07/onairplanebandwidthandlatency.html
George LeslieWaksman
<p>Having recently used Virgin America to transport myself across the
country, I was very pleased to have Internet access while I was in the
air. This, however, is not the sort of airplane bandwidth and latency
that I am going to talk about. Instead, I would like to discuss a
comparison between the bandwidth and latency of typical Internet
connections with those associated with taking a hard drive on an
airplane.</p>
<p>Let's say we compare a high speed (<span class="math">\(15\mathrm{Mbit}\)</span>) DSL connection to taking a
moderately large hard drive (<span class="math">\(500\mathrm{GB}\)</span>) on a plane for data rates between
San Francisco and Boston (<span class="math">\(\approx 7\mathrm{hr}\)</span>):</p>
<p>Bandwidth:</p>
<blockquote>
<ul class="simple">
<li><p>DSL: <span class="math">\(15\frac{\mathrm{Mbit}}{\mathrm{s}}\)</span></p></li>
<li><p>Airplane: <span class="math">\(\frac{500\mathrm{GB}}{7\mathrm{hr}} \times \frac{1\mathrm{hr}}{60\mathrm{min}} \times \frac{1\mathrm{min}}{60\mathrm{s}} \times \frac{8000\mathrm{Mbit}}{1\mathrm{GB}} \approx 150\frac{\mathrm{Mbit}}{\mathrm{s}}\)</span></p></li>
</ul>
</blockquote>
<p>Latency:</p>
<blockquote>
<ul class="simple">
<li><p>DSL: <span class="math">\(\approx 100\mathrm{ms}\)</span></p></li>
<li><p>Airplane: <span class="math">\(>7\mathrm{hr}\)</span></p></li>
</ul>
</blockquote>
<p>For fun, let's try something a little bigger on both sides: OC768 vs
Boeing 747400F plane filled with <span class="math">\(2\mathrm{TB}\)</span> hard drives.</p>
<p>Bandwidth:</p>
<blockquote>
<ul class="simple">
<li><p>OC768: <span class="math">\(38\frac{\mathrm{Gbit}}{\mathrm{s}}\)</span></p></li>
<li><p>747400F: <span class="math">\(\frac{250000\mathrm{lb}}{7\mathrm{hr}} \times \frac{2\mathrm{TB}}{1.7\mathrm{lb}} \times \frac{1\mathrm{hr}}{60\mathrm{min}} \times \frac{1\mathrm{min}}{60\mathrm{s}} \times \frac{8\mathrm{Tbit}}{1\mathrm{TB}} \approx 93\frac{\mathrm{Tbit}}{\mathrm{s}}\)</span></p></li>
</ul>
</blockquote>
<p>Latency:</p>
<blockquote>
<ul class="simple">
<li><p>OC768: <span class="math">\(<100\mathrm{ms}\)</span></p></li>
<li><p>747400F: <span class="math">\(>7\mathrm{hr}\)</span></p></li>
</ul>
</blockquote>
<p>Clearly, hard drives on an airplane will win in a purely bandwidth
driven application but airplanes suffer from incredibly high latency.
You will have to decide which is best choice based on your particular
use scenario.</p>
computers
intzorweb
https://gwax.com/blog/2009/07/onairplanebandwidthandlatency.html
Fri, 03 Jul 2009 00:21:27 GMT

From the land of ASCII
https://gwax.com/blog/2005/11/moreofthesame.html
George LeslieWaksman
<pre class="literalblock">And out of nowhere, 3D ASCII art:
,, ,
,/ ,/ ,' ,
,/  ,/  ,'  ,' 
+'  +' 
       
       
       
 *+  *+
 ,/  ,/  ,'  ,'
,/ ,/ ,' ,'
' '</pre>
<p>Yeah, I know it's not that great, but I was staring at a whole bunch of
text a bit ago and I started wondering if you could do half decent 3D
with ASCII art. Consider it an unpolished proofofconcept. Oh and if
it's turning up as a pile of gobbledygook, you need to change your
browsers monospace font to be an actual monospace font.</p>
computers
https://gwax.com/blog/2005/11/moreofthesame.html
Wed, 23 Nov 2005 08:22:24 GMT

5ud0 h4x0r1ng
https://gwax.com/blog/2005/08/5ud0h4x0r1ng.html
George LeslieWaksman
<div class="code"><pre class="code sh"><a id="rest_code_7d975769e6624879a79b8299d22f6bbc1" name="rest_code_7d975769e6624879a79b8299d22f6bbc1" href="https://gwax.com/blog/2005/08/5ud0h4x0r1ng.html#rest_code_7d975769e6624879a79b8299d22f6bbc1"></a>sudo<span class="w"> </span>su
</pre></div>
computers
https://gwax.com/blog/2005/08/5ud0h4x0r1ng.html
Tue, 09 Aug 2005 15:53:20 GMT

C100H202, 1ns: check
https://gwax.com/blog/2005/04/c100h2021nscheck.html
George LeslieWaksman
<p>My simulation code seems to be working pretty solidly now and I've
gotten almost my entire polyethylene model working. As of right now, I
haven't addressed bond eclipsing and some of my interaction constants
are a little off but, otherwise, everything seems qualitatively good.</p>
<p>Right now I've got my computer running a single
C<sub>100</sub>H<sub>202</sub> molecule at 273K. I am performing
calculations for every femtosecond, recording for every picosecond and
it will generate a full nanosecond in about 4 hours. After the
calculations are done, I'll render up images and make a movie, which I
will probably then post somewhere.</p>
<p>I'm a little disappointed in the state of the computation engine but I
don't care; I'm just so glad that I got the model working.</p>
<p><em>UPDATE:</em> It all worked out reasonably well and I have a video up on the
page that I just made up for my
<a class="reference external" href="https://gwax.com/thesis.html">thesis</a>. Sadly, initial
conditions and physics caused the system to develop a rotational mode
that, combined with some software limitations, makes most of the video
rather uninteresting. The first few seconds are nice. Further things
will be placed on that site as they appear and will, for the most part,
not be mentioned here.</p>
computers
school
https://gwax.com/blog/2005/04/c100h2021nscheck.html
Sat, 23 Apr 2005 03:59:33 GMT