Slashdot Mirror
Free Clock Democratizes Atomic Accuracy
schliz writes "A new, trial network of software-based clocks could give data centers and networks the accuracy of an atomic clock for free. The so-called RADclock analyses information from multiple computers across the internet by collecting the time from each machine's internal quartz clock, the time it takes for this information to be transmitted across the network, and comparing all the information collected to determine a time that is most likely to be accurate, so machines are calibrated across the network with up to microsecond accuracy — as good as that provided by a $50,000 atomic clock, researchers say."
"The accuracy of an atomic clock" decided by basically a median, a robot committee?
Well, you do get what you pay for. Not that it won't be awesome, I just don't think it will find applications where such accuracy is actually needed.
Real atomic clocks often have only have a nanosecond error. And new ones using ion gates are promising mobile clocks that are even more accurate. This is still pretty cool
refactor the law, its bloated, confusing and unmaintainable.
NTP solved this ages ago by distributing atomic clock accuracy through the network.
The only problem this will solve is where it is a private network not connected to public NTP servers (or organizations that do not trust public NTP). In that case, they would most likely be able to afford a atomic clock.
:. Ultimate Control Dedicated/VM Servers
With my super cool new software, i managed to get the first post!
I can imagine the speaking clock:
"At the third stroke, it will be, most likely, sixish"
They have atomic clocks on board and GPS receivers therefore give highly accurate time.
Donte Alistair Anderson Roberts - hi son!
Karma: Chameleon
Boy, that's going to leave a lot of nuclear scientists up in arms with a lot of unclaimed nuclear material on their hands.
...but in what situation would the time of day on a server or cluster need to be accurate down to a microsecond? Military, I would presume...but what else?
Living With a Nerd
I'm also pretty sure there are desktop clocks based on microcontrollers that implement ntp, so they display an accurate time without a computer.
Most data centers that really care about time nowadays install a commonly available GPS unit on site, which syncs clock time with all the atomic clocks in the flying GPS constellation.
Seriously, could the editor that greenlighted this have done a google search or something? It's getting embarrassing to read slashdot these days.
A long time ago (15 years!) I worked on the predecessor to http://www.symmetricom.com/products/gps-solutions/gps-time-frequency-receivers/XL-GPS/. I'm sure the modern day equivalent here doesn't cost 50k and will give you a local accurate time signal.
Donte Alistair Anderson Roberts - hi son!
Karma: Chameleon
Say I have 1000 PC's at my place of work and a certain percentage are usually a little fast (clock wise) while another percentage are usually a little slow.
If I use the average of all clocks to determine the "actual" time, wouldn't it slowly shift either to the slow side, or the fast side over time?
Say 550 of those 1000 are usually a second slow per day. That is a bias towards slow every time we take an average. We sync all clocks to the slow side and keep on repeating this. The longer you do this, the worse your accuracy is.
NTP works because we basically call one clock (or set of clocks that are synchronized) correct and we sync to those. If a correction to the master is needed, we will all correct at next sync.
Then again, for mental masturbation, how can you ever tell that one clock is right, versus another. Take two teams, put one on one side of the world and another team on the other. Give them a billion dollars each to develop the most precise clock in the world. When they are finished, compare the times on both. Which is correct? Especially considering time dilation, etc. I'll stop now since I've digressed considerably.
The Financial Sector.
Also, synchronized robotics, precisely coordinated CNC, and a host of other applications. Primarily, it's where absolute time isn't the concern, but rather where arbitrary time must be consistent between multiple devices (accounting for propagation delays, failures, etc...). Of course, protocols like PTP solve this fairly neatly: this particular product solves a different problem, and probably isn't actually useful.
There are two time issues to consider. One is how close your environment is to true time. The other is how close your individual devices are to one another. Messaging time-critical information between devices is severely complicated when the two devices are not on the same plane time-wise. Atomic clocks and the like solve the first problem. PTP solves the second problem. NTP almost (95%) solves both, but falls short in certain extremely time-critical situations.
GeekNights!
Late Night Radio for Geeks!
One of my favorite quotes relates to this;
Credit goes to Mark Twain (IIRC).
"When you have a watch/clock you always know what time it is. When you have 2 you are never quite sure."
"The price good men pay for indifference to public affairs is to be ruled by evil men." ~Plato (427-347 BC)
Can't wait to see all computers set their clocks to January 1st, 1980.
"A man with a watch knows what time it is. A man with two watches is never sure." -- Segal's Law.
What is the resolution of the built-in clock on most PCs? An Atomic clock might have nanoscale resolution, but if a computer's clock only has microsecond resolution, then it stands to reason that you can only synch the computer to within 1 microsecond of accuracy, no?
Something that can measure my sexual stamina properly ;-)
I want a list of atrocities done in your name - Recoil
What is the purpose of knowing time that accurately? You often want a clock signal that's that accurate, that's much more useful.
You can get 10MHz OCXO modules with a GPS receiver that syncs the OCXO to the satellites' on-board atomic clocks.
Much better.
So adding together a bunch of imprecise measurements together and getting the median gives you greater precision? No. Well, it might, maybe, but you won't know. depending on the statistical variance and how it is spread. But you won't know it is precise, you can only say how likely you are to be precise. that's not the same thing. Also, as I read it it collects and judges upon the time, whereafter it sets all the clocks to the median? Exactly how does this counteract drift? How long does it wait between updates? For how long does it collect samples? What they're doing is that they're collecting a number of samples, hoping that their variance is spread evenly, thus the median must be the precise time. Fun fact: that only increase the precision (and quite a bit!) if you collect an near-infinite amount of samples. I really do doubt they have a near-infinite amount of computer clocks.
This reminds me of the old Swiss watchmaker. Every day at noon, regular as clockwork, on office worker walked past the watchmaker's window on his way to lunch. This was the watchmaker's reminder to carry out his daily clock setting, so every time the office worker went past, the watchmaker checked that all of his clocks indicated noon.
One day the watchmaker happened to be out in the street at noon, and he mentioned this to the office worker. "That's funny", said the worker, "I always set my watch when I walk past your shop".
And so the network of software-based clocks will work fine, provided the computers from which the time is being aggregated are not themselves setting their time by this software-based clock.
Paid Q&A/Research
Thinking the IRS and your local state and community tax collectors would like to know everybody whose e-return is >= 1 microsecond late. That penalty money is important to g'ment - particularly in these days when an industrial base decimated by inequitable free trade provides ever less tax revenue.
Orwell: "In a Time of Universal Deceit, telling the Truth is a Revolutionary Act"
I think this piece of software might be useful for high speed (space) travel. Just read the theory of relativity, and you will know what I am talking about.
This sounds like real estate valuation scheme that led to the market crash. A Chinese mathematician said that you do not need to evaluate each property, just get an average of what the market says and you have your price. Then EVERYONE stopped doing proper evaluations and just looked at what EVERYONE else had for a price. If ALL the clocks switch to this, they ALL go off in some average direction.
Doing tests with 100 virtual machines does not show the effect. It must be with 1000 old computers to see it.
Bleah.. you don't know much about GPS timing, do you?
The message that the GPS satellite sends *includes the offset from UTC* including the number of leapseconds.
GPS time can't have leap seconds because it needs to be monotonic and continuous, like TAI time. UTC needs to line up with the Earth's rotation, so they need leap seconds.
Time arithmetic on GPS time is straightforward. Not so with UTC.
How does this account for rogue or malicious clocks present on the network? It sounds like it would be pretty easy to introduce significant error into the system.
I think I'll keep using one system connecting to the atomic clock and all the rest connecting to that one to keep their time accurate.
Edward Burr
Having a smoking section in a restaurant is like having a peeing section in a swimming pool.
The people behind this project tried to get the ntp hackers interested a year or two ago, but since the only thing they have is a possibly improved distributed algorithm, vs real NTP which has been designed to work even in the face of (intentional or accidental) falsetickers, nobody was very interested.
Any NTP wrangler who's the least interested in accurate local clocks will spend an hour and less than $100 to buy a cheap gps (Garmin GPS18(x) LVC) with Pulse Per Second (pps) output, a USB cable and a 9-pin RS232 connector:
Solder these together according to one of several howtos
(I can recommend http://www.satsignal.eu/ntp/FreeBSD-GPS-PPS.htm, it is quite similar to my own rooftop clock which I've connected to a FreeBSD host in my attic.)
and you'll have a clock source which on average will be exact, with a jitter of a microsecond or two.
Terje
(who used to compile and host the windows ntp binaries for a number of years)
"almost all programming can be viewed as an exercise in caching"
There was a story a few years back about a Security researcher that determined the quartz units in every computer are unique and have different enough time drift to fingerprint the individual machine's traffic despite IP address changes, proxies or anything similar. Botnets attacks wouldn't be tracked back to them, but CNC traffic to the botnets should still work.
Anyways this software comes out and samples millisecond differences between computer clocks on large networks, put this together with that guys Thesis work and you've got a worse end to privacy than Facebook.
Or something.
Has no one heard of Precision Time Protocol? http://en.wikipedia.org/wiki/Precision_time_protocol
The company I work at creates hardware that implements this protocol, we routinely get accuracies on the order of 10's of nanoseconds in our clocks. This is used by the military and in the telecom industry, for precise clock synchronization over IP/Ethernet.
This sounds like a Wikipedia approach to "truth"... you have enough people voting on it and sooner or later you have something that sort of resembles and might actually be truth. Or in this case, accuracy.
I would think there would be much better ways than this of obtaining accuracy. One of the basic problems with any voting scheme is that you can get a large amount of drift as it is assumed there is no standard other than concensus. NTP is a reference because there is are systems out there that are known to be anchored to accurate time standards.
I suppose if you were setting up a platform on Mars and had nothing else except concensus and voting this might be an approach. But that would seem to be about the only reason for accepting a scheme like this. On Earth there are way, way better ways of dealing with the problem that are actually rooted to accurate standards.
"Fetching the time via the web" is using NTP or SNTP to get time from better clocks. The article says that NTP is at best getting you millisecond accuracy, and claims RADclock can get to microsecond level.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
Open Source does not mean Free to use.
From the ptpd.sf.net README
- Legal notice -
PTPd was written by using only information contained within 'IEEE Std
1588-2008'. IEEE 1588 may contain patented technology, the use of which is not
under the control of the authors of PTPd. Users of IEEE 1588 may need to obtain
a license for the patented technology in the protocol. Contact the IEEE for
licensing information.
If only we had a publicly owned atomic clock, we could distribute time information using some sort of network protocol that compensates for latency and avoid all this fuss. We could even synchronize the time with satellites in orbit, and they, in turn, could transmit pulses at precise intervals.
I better write this down before someone else takes the credit.
https://www.eff.org/https-everywhere
Will they be connected with orbiting units or deep atmospheric ocean processors? Perhaps the space time will be redefined further and the stars will be closer to us than before.
you can download nist time, which will sync your computer to within a few msec of the NIST clock. when you do this, you will see that your pc is often off by hundreds of msecs the accuracy of nist time is limited by transmission errors across the web - how do they deal with this
This belongs in comp.compression
you had me at #!
From Wikipedia:
In August 2004, NIST scientists demonstrated a chip-scaled atomic clock.[8] According to the researchers, the clock was believed to be one-hundredth the size of any other. It was also claimed that it requires just 75 mW, making it suitable for battery-driven applications. This device could conceivably become a consumer product.
If atomic clocks become affordable and common, this might be a good idea. Otherwise, you're making up precision out of thin air, and you're probably better off with the ntpd approach wherein you have a hierarchy of timeservers that refer to atomic clocks, with the limited precision that comes with it. It's not like most systems need such precision, anyway. Why rush?
I suggest we use this type of approach to determine other values, such as fundamental physical constants. We could have all the computers of the world vote for what they think is the value of say... pi, or the number 2. One computer might think 2 is 1.9999999999999999999979 while another would report 2.000000000000000000001. By averaging these together, we should be able to get the exact floating point value for 2.
Let's just not pay attention to things like the difference between precision and accuracy anymore, it's too much work.
I mean, there's no way that the same physical limitations would apply to all quartz clocks, right?
Not clear why you would need nano second accuracy for computer work. But every GPS receiver provides that. GPS relies on knowing what time it is to within, I think, 10 ns. (A ns is about a 9" error. Non-differential GPS with WAIS is good to about 10 feet = about 13 ns
Third Career: Tree Farmer Second Career: Computer Geek First Career: Teacher, Outdoor Instructor, Photographer.