<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-family: Calibri, sans-serif; ">
<div>This is the value of building your own moderate scale cluster the first time.. Things like cable management take up a significant amount of time. And the cluster gods (anti-Grendel?) help you if you have a single bad cable in the rats nest to find.</div>
<div><br>
</div>
<div>You'll be a believer in color coding, number tags, etc. after hooking up a few dozen computers.</div>
<div><br>
</div>
<div>I've actually built a (very slow) cluster using WiFi interconnects (802.11a, as it happened). While it's wire free, it's also horribly slow, since everyone shares the same radio spectrum, so you have medium access contention issues, etc.. It could be
designed to be more efficient, say with free space optical interconnects. I think someone did a optical mesh of iPAQs once.</div>
<div><br>
</div>
<div>Power is trickier with USB than you think, because of the USB specs on how much power you can draw from a hub, etc. </div>
<div><br>
</div>
<div>Solar powered.. Now there's an interesting idea.. The Rpi takes 3.5W. At 1000W/square meter solar input, 20% conversion efficiency, etc. you get 200W/square meter. Or 57 pis/square meter. Or 175 square centimeter per pi.</div>
<div><br>
</div>
<div>The Rpi is about 9x6 cm (or 54 square cm).. So you need a good 4 times as much solar cell as Rpi, at least to run full tilt.</div>
<div><br>
</div>
<div>(as an aside.. The wireless cluster I mention above was part of a research project to assess the feasibility of a distributed computing solution for phased array antennas in space. The antenna element, the computing hardware, a battery, and solar cells
all stacked nicely.)</div>
<div><br>
</div>
<span id="OLK_SRC_BODY_SECTION">
<div style="font-family:Calibri; font-size:11pt; text-align:left; color:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt">
<span style="font-weight:bold">From: </span>Daniel Kidger <<a href="mailto:daniel.kidger@gmail.com">daniel.kidger@gmail.com</a>><br>
<span style="font-weight:bold">Date: </span>Tuesday, September 18, 2012 1:36 AM<br>
<span style="font-weight:bold">To: </span>"<a href="mailto:beowulf@beowulf.org">beowulf@beowulf.org</a>" <<a href="mailto:beowulf@beowulf.org">beowulf@beowulf.org</a>><br>
<span style="font-weight:bold">Subject: </span>[Beowulf] Southampton's RPi cluster is cool but too many cables?<br>
</div>
<div><br>
</div>
<div>
<div>All,
<div><br>
<div>Having seen the pictures of Simon's Raspberry Pi cluster at Southampton, one thing that strikes me is how ugly all the cables make it look.</div>
<div><br>
</div>
</div>
<div>So how can this be improved - indeed is 'cable-free' possible?</div>
<div>- The network could be Wifi using micro adapters. </div>
<div>- USB Power could be at least daisy-chained, but what about contactless as seen in various hip-bars for phone charging?</div>
<div>- Solar powered? Jim L: what is the maths here?</div>
<div><br>
</div>
<div>Daniel</div>
<div>Bull Information Systems</div>
</div>
</div>
</span>
</body>
</html>