Google NTP Servers and smearing leap seconds...
Time Technology and leaping seconds
http://googleblog.blogspot.com/2011/09/time-technology-and-leaping-seconds.html
"The solution we came up with came to be known as the “leap smear.” We
modified our internal NTP servers to gradually add a couple of
milliseconds to every update, varying over a time window before the
moment when the leap second actually happens. This meant that when it
became time to add an extra second at midnight, our clocks had already
taken this into account, by skewing the time over the course of the
day. All of our servers were then able to continue as normal with the
new year, blissfully unaware that a leap second had just occurred. We
plan to use this “leap smear” technique again in the future, when new
leap seconds are announced by the IERS."
It's interesting to see the title "Why time matters at Google" in the blog
when Androis bug 5485 has not been fixed in over two years, and most
Android-based phones are 15 seconds fast, as GPS-disciplined rather than
UTC. It appears only Samsung got around....
When hearing about this, Daniel Gambis at IERS told me that there was such
as strong momentum to drop leap seconds (ITU, etc...), yet the "damage" is
done !
On Thu, Sep 15, 2011 at 16:34, Doug Calvert dfc-list@douglasfcalvert.netwrote:
Time Technology and leaping seconds
http://googleblog.blogspot.com/2011/09/time-technology-and-leaping-seconds.html
"The solution we came up with came to be known as the “leap smear.” We
modified our internal NTP servers to gradually add a couple of
milliseconds to every update, varying over a time window before the
moment when the leap second actually happens. This meant that when it
became time to add an extra second at midnight, our clocks had already
taken this into account, by skewing the time over the course of the
day. All of our servers were then able to continue as normal with the
new year, blissfully unaware that a leap second had just occurred. We
plan to use this “leap smear” technique again in the future, when new
leap seconds are announced by the IERS."
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
That would be nice it is were a standard and everybody was doing it, because otherwise, for a day, your own network might be happy, but if you have any app that needs to communicate with the outside world, you are in a lot of hurt.
Didier KO4BB
Sent from my BlackBerry Wireless thingy while I do other things...
-----Original Message-----
From: Doug Calvert dfc-list@douglasfcalvert.net
Sender: time-nuts-bounces@febo.com
Date: Thu, 15 Sep 2011 16:34:39
To: Discussion of precise time and frequency measurementtime-nuts@febo.com
Reply-To: Discussion of precise time and frequency measurement
time-nuts@febo.com
Subject: [time-nuts] Google NTP Servers and smearing leap seconds...
Time Technology and leaping seconds
http://googleblog.blogspot.com/2011/09/time-technology-and-leaping-seconds.html
"The solution we came up with came to be known as the “leap smear.” We
modified our internal NTP servers to gradually add a couple of
milliseconds to every update, varying over a time window before the
moment when the leap second actually happens. This meant that when it
became time to add an extra second at midnight, our clocks had already
taken this into account, by skewing the time over the course of the
day. All of our servers were then able to continue as normal with the
new year, blissfully unaware that a leap second had just occurred. We
plan to use this “leap smear” technique again in the future, when new
leap seconds are announced by the IERS."
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
You can forget Wall St. firms and Banks for starters.
They need sub-microsecond accurate timing as some instruments (Forex)
are moving to <10 microsecond latency from order entry to order ack.
Although some software (FIXX) needs a lot of work to get there, they
are moving in that direction.
-George, N2FGX
Quoting shalimr9@gmail.com:
That would be nice it is were a standard and everybody was doing it,
because otherwise, for a day, your own network might be happy, but
if you have any app that needs to communicate with the outside
world, you are in a lot of hurt.
Didier KO4BB
Sent from my BlackBerry Wireless thingy while I do other things...
-----Original Message-----
From: Doug Calvert dfc-list@douglasfcalvert.net
Sender: time-nuts-bounces@febo.com
Date: Thu, 15 Sep 2011 16:34:39
To: Discussion of precise time and frequency measurementtime-nuts@febo.com
Reply-To: Discussion of precise time and frequency measurement
time-nuts@febo.com
Subject: [time-nuts] Google NTP Servers and smearing leap seconds...
Time Technology and leaping seconds
http://googleblog.blogspot.com/2011/09/time-technology-and-leaping-seconds.html
"The solution we came up with came to be known as the “leap smear.” We
modified our internal NTP servers to gradually add a couple of
milliseconds to every update, varying over a time window before the
moment when the leap second actually happens. This meant that when it
became time to add an extra second at midnight, our clocks had already
taken this into account, by skewing the time over the course of the
day. All of our servers were then able to continue as normal with the
new year, blissfully unaware that a leap second had just occurred. We
plan to use this “leap smear” technique again in the future, when new
leap seconds are announced by the IERS."
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
In message 20110916084529.80082q5qgwy5p1jd@host111.hostmonster.com, xaos@dark
smile.net writes:
And you can forget most physicists and metrologists as well, the do not
want wavelengths, half-lifes and energies to depend on the time of day...
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
xaos@darksmile.net said:
You can forget Wall St. firms and Banks for starters.
They need sub-microsecond accurate timing as some instruments (Forex) are
moving to <10 microsecond latency from order entry to order ack.
10 microsecond latency doesn't say anything about how accurate the time has
to be.
Does anybody have a good URL on the accuracy requirements of banks and/or
stock markets? I expect there are both legal and technical issues. I'd like
to understand them separately but I won't be surprised if they are thoroughly
tangled.
--
These are my opinions, not necessarily my employer's. I hate spam.
You are right.
To be more precise, I should have said that the time sync should be at
least 1 order of magnitude less. In the case of <10us turnaround time,
it is assumed that the timesync is <1us. This is the reason that
everyone uses multiple stratum 1 NTP servers using GPS in their
datacenters.
So the Forex transaction goes like this:
- (Both parties) Are we in proper sync timewise?
- (Party 1) I need transaction type x. My timestamp is: xxxx.xxxx. Go.
- (PArty 2) Confirmed. My timestamp is: yyyy.yyyy. Go.
These timestamps are legal entities and bind both parties to the transaction.
That's why transactions have a data transfer entity in the middle
(Reuters, Bloomberg) which guarantees proper timesync for all involved.
With Reuters in the middle, only the Reuters timestamp (arrival time
and send time) can be trusted.
However, Many times you will see a Reuters machine lose sync and the
UNIX SA's will restart NTP on it. Reuters puts more than one machine
per site for redundancy.
Quoting Hal Murray hmurray@megapathdsl.net:
xaos@darksmile.net said:
You can forget Wall St. firms and Banks for starters.
They need sub-microsecond accurate timing as some instruments (Forex) are
moving to <10 microsecond latency from order entry to order ack.
10 microsecond latency doesn't say anything about how accurate the time has
to be.
Does anybody have a good URL on the accuracy requirements of banks and/or
stock markets? I expect there are both legal and technical issues. I'd like
to understand them separately but I won't be surprised if they are thoroughly
tangled.
--
These are my opinions, not necessarily my employer's. I hate spam.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
I just read they were building a new transatlantic cable that will shave 10uS from the normal 60 or so uS and that for large traders, 1uS represents 100 million $ per year saving/increased revenue.
Didier KO4BB
Sent from my BlackBerry Wireless thingy while I do other things...
-----Original Message-----
From: xaos@darksmile.net
Sender: time-nuts-bounces@febo.com
Date: Fri, 16 Sep 2011 13:05:40
To: Discussion of precise time and frequency measurementtime-nuts@febo.com; Hal Murrayhmurray@megapathdsl.net
Reply-To: Discussion of precise time and frequency measurement
time-nuts@febo.com
Subject: Re: [time-nuts] Google NTP Servers and smearing leap seconds...
You are right.
To be more precise, I should have said that the time sync should be at
least 1 order of magnitude less. In the case of <10us turnaround time,
it is assumed that the timesync is <1us. This is the reason that
everyone uses multiple stratum 1 NTP servers using GPS in their
datacenters.
So the Forex transaction goes like this:
- (Both parties) Are we in proper sync timewise?
- (Party 1) I need transaction type x. My timestamp is: xxxx.xxxx. Go.
- (PArty 2) Confirmed. My timestamp is: yyyy.yyyy. Go.
These timestamps are legal entities and bind both parties to the transaction.
That's why transactions have a data transfer entity in the middle
(Reuters, Bloomberg) which guarantees proper timesync for all involved.
With Reuters in the middle, only the Reuters timestamp (arrival time
and send time) can be trusted.
However, Many times you will see a Reuters machine lose sync and the
UNIX SA's will restart NTP on it. Reuters puts more than one machine
per site for redundancy.
Quoting Hal Murray hmurray@megapathdsl.net:
xaos@darksmile.net said:
You can forget Wall St. firms and Banks for starters.
They need sub-microsecond accurate timing as some instruments (Forex) are
moving to <10 microsecond latency from order entry to order ack.
10 microsecond latency doesn't say anything about how accurate the time has
to be.
Does anybody have a good URL on the accuracy requirements of banks and/or
stock markets? I expect there are both legal and technical issues. I'd like
to understand them separately but I won't be surprised if they are thoroughly
tangled.
--
These are my opinions, not necessarily my employer's. I hate spam.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Should not be ms instead of us for a transatlantic cable? 60us at light
speed is only 18km ;)
Regards,
Javier
El 16/09/2011 23:53, shalimr9@gmail.com escribió:
I just read they were building a new transatlantic cable that will shave 10uS from the normal 60 or so uS and that for large traders, 1uS represents 100 million $ per year saving/increased revenue.
Didier KO4BB
Sent from my BlackBerry Wireless thingy while I do other things...
-----Original Message-----
From: xaos@darksmile.net
Sender: time-nuts-bounces@febo.com
Date: Fri, 16 Sep 2011 13:05:40
To: Discussion of precise time and frequency measurementtime-nuts@febo.com; Hal Murrayhmurray@megapathdsl.net
Reply-To: Discussion of precise time and frequency measurement
time-nuts@febo.com
Subject: Re: [time-nuts] Google NTP Servers and smearing leap seconds...
You are right.
To be more precise, I should have said that the time sync should be at
least 1 order of magnitude less. In the case of<10us turnaround time,
it is assumed that the timesync is<1us. This is the reason that
everyone uses multiple stratum 1 NTP servers using GPS in their
datacenters.
So the Forex transaction goes like this:
- (Both parties) Are we in proper sync timewise?
- (Party 1) I need transaction type x. My timestamp is: xxxx.xxxx. Go.
- (PArty 2) Confirmed. My timestamp is: yyyy.yyyy. Go.
These timestamps are legal entities and bind both parties to the transaction.
That's why transactions have a data transfer entity in the middle
(Reuters, Bloomberg) which guarantees proper timesync for all involved.
With Reuters in the middle, only the Reuters timestamp (arrival time
and send time) can be trusted.
However, Many times you will see a Reuters machine lose sync and the
UNIX SA's will restart NTP on it. Reuters puts more than one machine
per site for redundancy.
Quoting Hal Murrayhmurray@megapathdsl.net:
xaos@darksmile.net said:
You can forget Wall St. firms and Banks for starters.
They need sub-microsecond accurate timing as some instruments (Forex) are
moving to<10 microsecond latency from order entry to order ack.
10 microsecond latency doesn't say anything about how accurate the time has
to be.
Does anybody have a good URL on the accuracy requirements of banks and/or
stock markets? I expect there are both legal and technical issues. I'd like
to understand them separately but I won't be surprised if they are thoroughly
tangled.
--
These are my opinions, not necessarily my employer's. I hate spam.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
With the recent wild fluctuations in the Commodities markets it is
incredibly important that time delay between any 2 parties (in a
financial transaction) be reduced to a minimum.
Only a few years ago, gold, silver and oil would vary by a few dollars
(at most), intraday and even intraweek.
Lately, we had jumps of a few hundred dollars in a few minutes/seconds.
If a bank makes a few 100+ Mil transactions a day (trust me this is
not even a blip on the radar in some places) those extra microsecs add
up.
Now, one would think mathematically about this and say "it should
average out". Well, no! The market (for some weird reason) has a mind
of it's own and when you start to lose money, it just gets worse and
worse.
Even the best mathematical predictors (computer models and humans)
break down when the market volatility (randomness) goes up.
What does "up" or "down" volatility mean? Depends on the day.
The point is, when the proverbial shit hits the fan, you want to bail
out of your position ASAP. However, the time delay between you and
your overseas party added another 50-1000 microseconds to the closing
time, you just lost another xxxx dollars and in a losing day that adds
up!
I've worked on trading floors for many years and seen people drop dead
after they got closing confirmation!
No wonder most traders are so young.
Have a nice weekend everyone!
Quoting shalimr9@gmail.com:
I just read they were building a new transatlantic cable that will
shave 10uS from the normal 60 or so uS and that for large traders,
1uS represents 100 million $ per year saving/increased revenue.
Didier KO4BB
Sent from my BlackBerry Wireless thingy while I do other things...
-----Original Message-----
From: xaos@darksmile.net
Sender: time-nuts-bounces@febo.com
Date: Fri, 16 Sep 2011 13:05:40
To: Discussion of precise time and frequency
measurementtime-nuts@febo.com; Hal Murrayhmurray@megapathdsl.net
Reply-To: Discussion of precise time and frequency measurement
time-nuts@febo.com
Subject: Re: [time-nuts] Google NTP Servers and smearing leap seconds...
You are right.
To be more precise, I should have said that the time sync should be at
least 1 order of magnitude less. In the case of <10us turnaround time,
it is assumed that the timesync is <1us. This is the reason that
everyone uses multiple stratum 1 NTP servers using GPS in their
datacenters.
So the Forex transaction goes like this:
- (Both parties) Are we in proper sync timewise?
- (Party 1) I need transaction type x. My timestamp is: xxxx.xxxx. Go.
- (PArty 2) Confirmed. My timestamp is: yyyy.yyyy. Go.
These timestamps are legal entities and bind both parties to the transaction.
That's why transactions have a data transfer entity in the middle
(Reuters, Bloomberg) which guarantees proper timesync for all involved.
With Reuters in the middle, only the Reuters timestamp (arrival time
and send time) can be trusted.
However, Many times you will see a Reuters machine lose sync and the
UNIX SA's will restart NTP on it. Reuters puts more than one machine
per site for redundancy.
Quoting Hal Murray hmurray@megapathdsl.net:
xaos@darksmile.net said:
You can forget Wall St. firms and Banks for starters.
They need sub-microsecond accurate timing as some instruments (Forex) are
moving to <10 microsecond latency from order entry to order ack.
10 microsecond latency doesn't say anything about how accurate the time has
to be.
Does anybody have a good URL on the accuracy requirements of banks and/or
stock markets? I expect there are both legal and technical issues.
I'd like
to understand them separately but I won't be surprised if they are
thoroughly
tangled.
--
These are my opinions, not necessarily my employer's. I hate spam.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.