Archive for the ‘ntp server’ Category

Google Finds Innovative Way to Avoid Leap Seconds

Wednesday, September 28th, 2011

Leap Seconds have been in use since the development of atomic clocks and the introduction of the global timescale UTC (Coordinated Universal Time). Leap Seconds prevent the actual time as told by atomic clocks and the physical time, governed by the sun being highest at noon, from drifting apart.

Since UTC began in the 1970’s when UTC was introduced, 24 Leap Seconds have been added. Leap seconds are a point of controversy, but without them, the day would slowly drift into night (albeit after many centuries); however, they do cause problems for some technologies.

NTP servers (Network Time Protocol) implement Leap Seconds by repeating the final second of the day when a Leap Second is introduced. While Leap Second introduction is a rare event, occurring only once or twice a year, for some complex systems that process thousands of events a second this repetition causes problems.

For search engine giants, Google, Leap Seconds can lead to their systems from working during this second, such as in 2005 when some of its clustered systems stopped accepting work. While this didn’t lead to their site from going down, Google wanted to address the problem to prevent any future problems caused by this chronological fudge.

Its solution was to write a program that essentially lied to their computer servers during the day of a Leap Second, making the systems believe the time was slightly ahead of what the NTP servers were telling it.

This gradual speeding up time meant that at the end of a day, when a Leap Second is added, Google’s timeservers do not have to repeat the extra second as the time on its servers would already be a second behind by that point.

Galleon GPS NTP server

Whilst Google’s solution to the Leap Second is ingenious, for most computer systems Leap Seconds cause no problems at all. With a computer network synchronised with an NTP server, Leap Seconds are adjusted automatically at the end of a day and occur only rarely, so most computer systems never notice this small hiccup in time.

 

Oddities of Time and the Importance of Accuracy

Wednesday, September 14th, 2011

Most of us think we know what the time it is. At a glance of our wristwatches or wall clocks, we can tell what time it is. We also think we have a pretty good idea of the speed time move forwards, a second, a minute, an hour or a day are pretty well-defined; however, these units of time are completely man-made and are not as constant as we may think.

Time is an abstract concept, while we may think it is the same for everybody, time is affected by its interaction with the universe. Gravity, for instance, as Einstein observed, has the ability to warp space-time altering the speed in which time passes, and while we all live on the same planet, under the same gravitational forces, there are subtle differences in the speed in which time passes.

Using atomic clocks, scientists are able to establish the effect Earth’s gravity has on time. The high above sea level an atomic clock is placed, the quicker time travels. While these differences are minute, these experiments clearly demonstrate that Einstein’s postulations were correct.

Atomic clocks have been used to demonstrate some of Einstein’s other theories regarding time too. In his theories of relativity, Einstein argued that speed is another factor that affects the speed at which time passes. By placing atomic clocks on orbiting spacecraft or aeroplanes travelling at speed, the time measured by these clocks differs to clocks left static on Earth, another indication that Einstein was right.

Before atomic clocks, measuring time to such degrees of accuracy was impossible, but since their invention in the 1950’s, not only have Einstein’s postulations proved right, but also we have discovered some other unusual aspects to how we regard time.

While most of us think of a day as 24-hours, with every day being the same length, atomic clocks have shown that each day varies. Furthermore, atomic clocks have also shown that the Earth’s rotation is gradually slowing down, meaning that days are getting slowly longer.

Because of these changes to time, the world’s global timescale, UTC (Coordinated Universal Time) needs occasional adjustments. Every six months or so, leap seconds are added to ensure UTC runs at the same rate as an Earth day, accounting for the gradual slowing down of the planet’s spin.

For technologies that require high levels of accuracy, these regular adjustments of time are accounted for by the time protocol NTP (Network Time Protocol) so a computer network using an NTP time server is always kept true to UTC.

British Atomic Clock Leads Race for Accuracy

Friday, September 2nd, 2011

NPL's atomic clock

Researchers have discovered that the British atomic clock controlled by the UK’s National Physical Laboratory (NPL) is the most accurate in the world.

NPL’s CsF2 caesium fountain atomic clock is so accurate that it wouldn’t drift by a second in 138 million years, nearly twice as accurate as first thought.

Researchers have now discovered the clock is accurate to one part in 4,300,000,000,000,000 making it the most accurate atomic clock in the world.

The CsF2 clock uses the energy state of caesium atoms to keep time. With a frequency of 9,192,631,770 peaks and troughs every second, this resonance now governs the international standard for an official second.

The international standard of time—UTC—is governed by six atomic clocks, including the CsF2, two clocks in France, one in Germany and one in the USA, so this unexpected increase in accuracy means the global timescale is even more reliable than first thought.

UTC is essential for modern technologies, especially with so much global communication and trade being conducted across the internet, across borders, and across timezones.

UTC enables separate computer networks in different parts of the world to keep exactly the same time, and because of its importance accuracy and precision is essential, especially when you consider the types of transactions now conducted online, such as the buying of stocks and shares and global banking.

Receiving UTC requires the use of a time server and the protocol NTP (Network Time Protocol). Time servers receive a source of UTC direct from atomic clocks sources such as NPL, who broadcast a time signal over long wave radio, and the GPS network (GPS satellites all transmit atomic clock time signals, which is how satellite navigation systems calculate position by working out the difference in time between multiple GPS signals.)

NTP keeps all computers accurate to UTC by continuously checking each system clock and adjusting for any drift compared to the UTC time signal. By using an NTP time server, a network of computers is able to remain within a few milliseconds of UTC preventing any errors, ensuring security and providing an attestable source of accurate time.

 

What Governs our Clocks

Tuesday, August 23rd, 2011

Most of us recognise how long an hour, a minute, or a second is, and we are used to seeing our clocks tick past these increments, but have you ever thought what governs clocks, watches and the time on our computers to ensure that a second is a second and an hour an hour?

Early clocks had a very visible form of clock precision, the pendulum. Galileo Galilei was the first to discover the effects of weight suspended from a pivot. On observing a swinging chandelier, Galileo realised that a pendulum oscillated continuously above its equilibrium and didn’t falter in the time between swings (although the effect weakens, with the pendulum swinging less far, and eventually stops) and that a pendulum could provide a method of keeping time.

Early mechanical clocks that had pendulums fitted proved highly accurate compared to other methods tried, with a second able to be calibrated by the length of a pendulum.

Of course, minute inaccuracies in measurement and effects of temperature and humidity meant that pendulums were not wholly precise and pendulum clocks would drift by as much as half an hour a day.

The next big step in keeping track of time was the electronic clock. These devices used a crystal, commonly quartz, which when introduced to electricity, will resonate. This resonance is highly precise which made electric clocks far more accurate than their mechanical predecessors were.

True accuracy, however, wasn’t reached until the development of the atomic clock. Rather than using a mechanical form, as with a pendulum, or an electrical resonance as with quartz, atomic clocks use the resonance of atoms themselves, a resonance that doesn’t change, alter, slow or become affected by the environment.

In fact, the International System of Units that define world measurements, now define a second as the 9,192,631,770 oscillations of a caesium atom.

Because of the accuracy and precision of atomic clocks, they provide the source of time for many technologies, including computer networks. While atomic clocks only exist in laboratories and satellites, using devices like Galleon’s NTS 6001 NTP time server.

A time server such as the NTS 6001 receives a source of atomic clock time from either GPS satellites (which use them to provide our sat navs with a way to calculate position) or from radio signals broadcast by physics laboratories such as NIST (National Institute of Standards and Time) or NPL (National Physical Laboratory).

 

Receiving Time Signals with GPS

Wednesday, August 17th, 2011

Accurate time is one of the most important aspects to keeping a computer network secure and safe. Places such as stock exchanges, banks and air traffic control rely on secure and accurate time. As computers rely on time as their only reference for when events happen, a slight error in a time code could lead to all sorts of errors, from millions being wiped off share prices to aeroplane flight paths being incorrect.

GPS time signals are secure

And time doesn’t just need to be accurate for these organizations, but secure too. A malicious user who interferes with a timestamp could cause all sorts of trouble, so ensuring time sources are both secure and accurate is vital.

Security is increasingly important for all sorts of organisations. With so much trade and communication conducted over the internet, using a source of accurate and secure time is as important a part of network security as anti-virus and firewall protection.

Despite the need for accuracy and security, many computer networks still rely on online time servers. Internet sources of time are not only unreliable, with inaccuracies commonplace, and distance and latency affecting the precision, but an Internet time server is also unsecure and able to be hijacked by malicious users.

But an accurate, reliable and completely secure source of time is available everywhere, 365 days a year—GPS.

While commonly thought of as a means of navigation, GPS actually provides an atomic clock time code, direct from the satellite signals. It is this time code that navigation systems use for calculating position but it is just as effective to provide a secure time stamp for a computer network.

Organizations that rely on accurate time for safety and security all use GPS, as it is a continuous signal, that never goes down, is always accurate and can’t be interfered with by third parties.

To utilise GPS as a source of time, all that is required is a GPS time server. Using an antenna, the time server receives the GPS signal, while NTP (Network Time Protocol) distributes it around the network.

GPS time server

With a GPS time server, a computer network is able to maintain accuracy to within a few milliseconds of the atomic clock time signal, which is translated into UTC time (Coordinated Universal Time) thanks to NTP, ensuring the network is running the same accurate time as other networks also synchronised to a UTC time source.

Precise Time on the Markets

Wednesday, August 10th, 2011

The stock market has been in the news a lot lately. As global uncertainty about national debts rise, the markets are in flux, with prices changing incredibly quickly. On a trading floor, every second counts and precise time is essential for global buying and selling of commodities, bonds and shares.

NTS 6001 from Galleon Systems

The international stock exchanges such as the NASDAQ and London Stock Exchange all require accurate and precise time. With traders buying and selling shares for customers across the globe, a few seconds of inaccuracy could cost millions as share prices fluctuate.

NTP servers linked to atomic clock timing signals ensure that the stock exchange keeps an accurate and precise time. As computers across the globe all receive the stock prices, as and when they change, these two use NTP server systems to maintain time.

The global timescale UTC (Coordinated Universal Time) is used as the basis for atomic clock timing, so no matter where a trader is on the globe, the same timescale prevents confusion and errors when dealing with stocks and shares.

Because of the billions of pounds worth of stocks and shares that are bought and sold on trading floors every day, security is essential. NTP servers work externally to networks, getting their time from sources such as GPS (Global Positioning System) or radio signals put out by organisations like the National Physical Laboratory (NPL) or the National Institute for Standards and Time (NIST).

The stock exchanges can’t use a source of internet because of the risk this could pose. Hackers and malicious users could tamper with the time source, leading to mayhem and cost millions and perhaps billions if the wrong time was spread around the exchanges.

The precision of internet time is limited too. Latency over distance can create delays, which could lead to errors, and if the time source ever went down, the stock markets could hit trouble.

It is not only stock markets that need precise and accurate time, computer networks across the globe concerned about security use dedicated NTP servers like Galleon Systems’ NTS 6001. Providing accurate time from both GPS and radio signals from NPL and NIST, the NTS 6001 ensure accurate, precise and secure time every day of the year.

 

Hackers and Time Servers

Wednesday, August 3rd, 2011

Computer hacking is a common subject in the news. Some of the biggest companies have fallen victim to hackers, and for a myriad of reasons. Protecting computer networks from invasion from malicious users is an expensive and sophisticated industry as hackers use many methods to invade a system.

Various forms of security exist to defend against unauthorised access to computer networks such as antivirus software and firewalls.

One area often overlooked, however, is where a computer network gets it source of time from, which can often be a vulnerable aspect to a network and a way in for hackers.

NTP servers helps secure computer networks

Most computer networks use NTP (Network Time Protocol) as a method of keeping synchronised. NTP is excellent at keeping computers at the same time, often to within a few milliseconds, but is dependent on a single source of time.

Because computer networks from different organisations need to communicate together, having the same source of time makes sense, which is the reason most computer networks synchronise to a source of UTC (Coordinated Universal Time).

UTC, the world’s global timescale, is kept true by atomic clocks and various methods of utilising UTC are available.

Quite often, computer networks use an internet time source to obtain UTC but this is often when they run into security issues.

Using internet time sources leave a computer network open to several vulnerabilities. Firstly, to allow access to the internet time source, a port needs keeping open in the system firewall (UDP 123). As with any open port, unauthorised users could take advantage of this, using the open port as a way into the network.

Secondly, if the internet time source itself if tampered with, such as by BGP injection (Border Gateway Protocol) this could lead to all sorts of problems. By telling internet time servers it was a different time or date, major havoc could ensue with data getting lost, system crashes—a type of Y2K effect!

Finally, internet time servers can’t be authenticated by NTP and can also be inaccurate. Vulnerable to latency and affected to distance, errors can also occur; earlier this year some reputable time servers lost several minutes, leading to thousands of computer networks receiving the wrong time.

To ensure complete protection, dedicated and external time servers, such as Galleon’s NTS 6001 are the only secure method of receiving UTC. Using GPS (or a radio transmission) an external NTP time server can’t be manipulated by malicious users, is accurate to a few milliseconds, can’t drift and is not susceptible to timing errors.

 

75 Years of the Speaking Clock

Wednesday, July 27th, 2011

Britain’s speaking clock celebrates its 75th birthday this week, with the service still providing the time to over 30 million callers a year.

75 Years of the speaking clock

The service, available by dialling 123 on any BT landline (British Telecom), began in 1936 when the General Post Office (GPO) controlled the telephone network. Back then, most people used mechanical clocks, which were prone to drift. Today, despite the prevalence of digital clocks, mobile phones, computers and a myriad number of other devices, the BT speaking clock still provides the time to 30 million callers a year, and other networks implement their own speaking clock systems.

Much of the speaking clock’s continuing success is perhaps down to the accuracy that it keeps. The modern speaking clock is accurate to five milliseconds (5/1000ths of a second), and kept precise by the atomic clock signals provided by NPL (National Physical Laboratory) and the GPS network.

But the announcer declaring the time ‘after the third stroke’ provides people with a human voice, something other time-telling methods don’t provide, and may have something to do with why so many people still use it.

Four people have had the honour of providing the voice for the speaking clock; the current voice of the BT clock is Sara Mendes da Costa, who has provided the voice since 2007.

Of course, many modern technologies require an accurate source of time. Computer networks that need to keep synchronised, for security reasons and to prevent of errors, require a source of atomic clock time.

Network time servers, commonly called NTP servers after Network Time Protocol that distributes the time across the computers on a network, use either GPS signals, which contain atomic clock time signals, or by radio signals broadcast by places like NPL and NIST (National Institute for Standards and Time) in the US.

 

Clock to Run for 10,000 Years

Wednesday, July 20th, 2011

The construction of clock, designed to tell the time for 10,000 years, is underway in Texas. The clock, when built, will stand over 60 metres tall and will have a clock face nearly three metres across.

Built by a non-profit organisation, the Long Now Foundation, the clock is being built so as to, not only still be standing in 10,000 years, but also still be telling the time.

A model of the Long Now Clock, which is expected to run for 10,000 years

Consisting of a 300kg gear wheel and a 140kg steel pendulum, the clock will tick every ten seconds and will feature a chime system that will allow 3.65 million unique chime variations—enough for 10,000 years of use.

Inspired by ancient engineering projects of the past, such as the Great Wall of China and the Pyramids—objects designed to last, the clock’s mechanism will feature state-of-the-art materials that don’t require lubrication of servicing.

However, being an mechanical clock, the Long Now Clock will not be very accurate and will require resetting to avoid drift otherwise the time in 10,000 years will not represent the time on Earth.

Even atomic clocks, the world’s most accurate clocks, require help in preventing drift, not because the clocks themselves drift—atomic clocks can remain accurate to a second for 100 million years, but the Earth’s rotation is slowing.

Every few years an extra second is added to a day. These Leap Seconds inserted on to UTC (Coordinated Universal Time) prevent the timescale and the movement of the Earth from drifting apart.

UTC is the global timescale that governs all modern technologies from satellite navigation systems, air traffic control and even computer networks.

While atomic clocks are expensive laboratory-based machines, receiving the time from an atomic clock is simple, requiring only a NTP time server (Network Time Protocol) that uses either GPs or radio frequencies to pick up time signals distributed by atomic clock sources. Installed on a network, and NTP time server can keep devices running to within a few milliseconds of each other and of UTC.

 

 

How Long is a Day?

Wednesday, July 13th, 2011

A day is something most of us take for granted, but the length of a day is not as simple as we may think.

A day, as most of us know, is the time it takes for the Earth to spin on its axis. Earth takes 24 hours to do one complete revolution, but other planets in our solar system have day lengths far different to ours.

Galleon NTS 6001

The largest planet, Jupiter, for instance, takes less than ten hours to spin a revolution making a Jovian day less than half of that of Earth, while a day on Venus is longer than its year with a Venusian day 224 Earth days.

And if you think of those plucky astronauts on the international Space Station, hurtling around the Earth at over 17,000 mph, a day for them is just 90 minutes long.

Of course, few of us will ever experience a day in space or on another planet, but the 24-hour day we take for granted is not as steadfast as you may think.

Several influences govern the revolution of the Earth, such as the movement of tidal forces and the effect of the Moon’s gravity. Millions of years ago, the Moon was much closer to Earth as it is now, which caused much higher tides, as a consequence the length of Earth’s day was shorter—just 22.5 hours during the time of the dinosaurs. And ever since the earth has been slowing.

When atomic clocks were first developed in the 1950’s, it was noticed that the length of a day varied. With the introduction of atomic time, and then Coordinated Universal Time (UTC), it became apparent that the length of a day was gradually lengthening. While this change is very minute, chorologists decided that to ensure equilibrium of UTC and the actual time on Earth—noon signifying when the sun is at its highest above the meridian—additional seconds needed to be added, once or twice a year.

So far, 24 of these ‘Leap Seconds’ have been since 1972 when UTC first became the international timescale.

Most technologies dependent on UTC use NTP servers like Galleon’s NTS 6001, which receives accurate atomic clock time from GPS satellites. With an NTP time server, automatic leap second calculations are done by the hardware ensuring all devices are kept accurate and precise to UTC.