<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body>
<p>After missing the opportunity to "soak test" our Starlink RV kit
in the Auckland floods the other week, Cyclone Gabrielle (which is
still noticeably upon us with very strong winds) was an
opportunity I couldn't let pass. I've learned a lot over the last
day or so, and it's been interesting to say the least. <br>
</p>
<p>So I put Dishy on our deck (slightly obscured view of the
southern sky - the only bit of sky it will voluntarily look at for
any length of time here in NZ) and tied it down on my painter's
trestle with a few cable ties - so far it hasn't blown away. <br>
</p>
<p>I set up a ping series with 86400 large packets (1448 bytes to
mimic you-know-what) to one of our uni machines to last for a day,
starting just before the worst rain was forecast. I also put a
beaker down next to it for some rainfall measurements - it
collected 152 mm (~6") over the duration of the ping experiment.
Not as bad as the deluge by a far shot but still, decent rain.
Plus a lot of other areas got heaps more, and with Starlink,
gateway and fellow users taking on water also impact on what you
can do.<br>
</p>
<p>I also ran a few speedtest.net tests to a server at a local ISP
at different times during the rain yesterday afternoon and last
night.</p>
<p>Note I used the Starlink Ethernet adapter for this, not the WiFi
from the router. <br>
</p>
<p>Here is what I got in terms of average and minimum RTTs and
packet loss over a 60 s interval. Times approximate in local time:</p>
<img src="cid:part1.mUH558O8.1GRhi0OC@auckland.ac.nz" moz-do-not-send="false">
<p> Note:</p>
<ul>
<li>High minimum RTTs to start with (higher than for the
speedtest.net tests, possibly because the packets are larger,
possibly because Starlink games speedtest.net, possibly because
of congestion thanks to everyone having been told to work from
home that day and schools closing).</li>
<li>Noticeable jumps in RTT at around 9:30 pm and just before
midnight, and then again at around 3:30 and 4:45 am and some
time before 7 am. There were A LOT of power outages - not at our
place (and we're USP'd with a 10 kWh solar battery anyway) but
around ALL three Starlink gateways around the larger Auckland
region (Clevedon, Te Hana and Puwera). I'm unsure whether it's a
result of a large number of lifestyle block belters (the only
serious Starlink customer group around here) losing power and
thus freeing up capacity, or whether it's some change in the
gateway arrangements. <br>
</li>
<li>Longest continuous ping fail: 22s</li>
<li>Packet loss appears elevated during the height of the rain.
Rain fade?<br>
</li>
</ul>
<p>Speedtest.net downloads didn't hit or exceed the 100 Mb/s mark
until about 11 pm, and some runs yielded as little as 6.5 Mb/s
down and 1.49 Mb/s up. Average down was all over the place,
average up typically in the mid single digits.<br>
</p>
<p>On the pro side, Starlink basically remained usable throughout,
certainly for someone wanting to follow what was happening.<br>
</p>
<pre class="moz-signature" cols="72">--
****************************************************************
Dr. Ulrich Speidel
School of Computer Science
Room 303S.594 (City Campus)
The University of Auckland
<a class="moz-txt-link-abbreviated" href="mailto:u.speidel@auckland.ac.nz">u.speidel@auckland.ac.nz</a>
<a class="moz-txt-link-freetext" href="http://www.cs.auckland.ac.nz/~ulrich/">http://www.cs.auckland.ac.nz/~ulrich/</a>
****************************************************************
</pre>
</body>
</html>