I'm no NTP expert so I'm not sure what it does differently, but the more detailed paper is at <a href="http://research.google.com/archive/spanner.html">http://research.google.com/archive/spanner.html</a><div><br></div>
<div>Eric<br><br><div class="gmail_quote">On Mon, Nov 26, 2012 at 12:40 PM, Russ Allbery <span dir="ltr"><<a href="mailto:rra@stanford.edu" target="_blank">rra@stanford.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="HOEnZb"><div class="h5">Ted Cabeen <<a href="mailto:ted@cabeen.org">ted@cabeen.org</a>> writes:<br>
<br>
> Interesting article on how Google has moved beyond NTP, with a mention<br>
> of last year's leap second:<br>
> <a href="http://www.wired.com/wiredenterprise/2012/11/google-spanner-time/" target="_blank">http://www.wired.com/wiredenterprise/2012/11/google-spanner-time/</a><br>
<br>
</div></div>It would be interesting to have more details, since that article makes a<br>
lot of Google not using NTP and then proceeds to describe a network<br>
protocol that sounds exactly like NTP, down to how it distributes stratum<br>
zero clock sources to other systems in the same data center. So it's not<br>
clear exactly what they changed.<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
Russ Allbery (<a href="mailto:rra@stanford.edu">rra@stanford.edu</a>) <<a href="http://www.eyrie.org/~eagle/" target="_blank">http://www.eyrie.org/~eagle/</a>><br>
</font></span></blockquote></div><br></div>