At 3:29 PM +0200 2005-09-13, Brad Knowles wrote:
You're talking about building an OLTP database the likes of which the
world has never seen before, and putting that into each and every NTP
server in the world.
Okay, maybe just every pool server, but that doesn't change the
scale of the problem at the host level.
Yes, big honking Oracle RAS servers may be able to
get more than a few hundred queries per second, but not with jitter and
latency measured to such extremely low levels. To make this useful for a
time server, you'd have to be down into the single millisecond range, or
even into the nanoseconds, for both latency and jitter.
And no OLTP database in the world has ever attempted that.
I know some people who are in the process of implementing an OLTP
server to handle hundreds, thousands, tens of thousands of queries
per second, and with latencies down into the 100ms range. This is a
commercial mega-buck product, and they've told me that this problem
is of a scale such that they had never seen before.
They think that they've finally gotten a handle on the problem at
this level, but no one has any ideas yet on how they're going to
approach the customers who are asking for latencies into the five
millisecond range.
If you want low latency/low jitter (i.e., down into the
single-digit millisecond range), you have to store the data in
memory. There's just no way around that problem, at least not with
the latest technology we have available to us today.
--
Brad Knowles, <[EMAIL PROTECTED]>
"Those who would give up essential Liberty, to purchase a little
temporary Safety, deserve neither Liberty nor Safety."
-- Benjamin Franklin (1706-1790), reply of the Pennsylvania
Assembly to the Governor, November 11, 1755
SAGE member since 1995. See <http://www.sage.org/> for more info.
_______________________________________________
timekeepers mailing list
[email protected]
https://fortytwo.ch/mailman/cgi-bin/listinfo/timekeepers
