A Look Behind The Scenes At The Tour De France With Dimension Data

Tour De France

The Tour de France first took place in 1903 and has gained global prominence and popularity ever since. In recent years a partnership with Dimension Data has revolutionised the race coverage, making it one of the most interactive sports events in the world, on both sides of the camera. Russell Drury finds out more about the technology that keeps the race rolling.

French media group Amaury Sport Organisation (ASO) owns a variety of high profile sporting events around the globe, including a number of professional cycling races, which includes the iconic Tour de France.

Prior to 2015, TV commentators on the Tour de France, as well as viewers and fans following the race, had limited access to real time information about the riders’ performance, and were only able to track the entire peloton rather than an individual rider. ASO wanted to deliver a much richer experience to the billions of cycling fans around the world by providing greater access to existing race data, as well as by sharing riders’ individual locations and speeds with audiences. To bring the race into the digital era, ASO was also keen to make sure this data wasn’t just available on TV, but also online and via social media channels.

GPS tracking devices were fitted to the saddles of each of the Tour’s 198 riders’ bikes
GPS tracking devices were fitted to the saddles of each of the Tour’s 198 riders’ bikes

In 2015, prior to the start of the Tour, ASO signed a five-year Official Technology Partner agreement with Dimension Data in order to transform the viewing experience of fans, race commentators and the general media. At the centre of this transformation was the real time tracking and data analytics platform that Dimension Data designed, built and implemented in just four months, with teams collaborating across 11 cities across five continents in order to build and test the solution. The ability to know where the riders are at any point in time during the race, the composition of the peloton, obtaining real time information about the speeds of the riders, and the distance between them, was the first step to providing a new viewing experience for professional cycling race commentators, fans, and viewers.

To make this possible, GPS tracking devices were fitted to the saddles of each of the 198 riders’ bikes. These sensors transmitted data on rider speed, GPS coordinates as well as other information, all in real time, producing between four and six million data records during each of the 21 stages of the Tour. The data produced by the devices was relayed to aircraft following the race, and from there to Dimension Data’s Big Data Truck – a fully mobile data centre – located in the technical zone near the finish line of each stage. The Big Data Truck was connected to the Dimension Data cloud platform, where the data was analysed through an enterprise grade data analytics solution, and then served up on a live tracking website and as television graphics on screens around the world to a cumulative TV audience of 3.5 billion people.

Hybrid cloud

The hybrid cloud environment used for the Tour de France combines both modular on site equipment with multiple off-premise cloud services, delivering redundancy and near limitless scalability during the race. The design is based on three cornerstones:

• High availability and security – ensuring no single point of failure and protecting the data from intrusion or leakage
• A diversity of locations – along with the Big Data Truck, Dimension Data hosts redundant services in its data centres in London and Amsterdam
• A diversity of vendors and smart architecture – designed to be robust enough to handle any eventuality during the race, including inclement weather and rider crashes

Every day after the end of each stage of the race, Dimension Data’s Big Data Truck was driven to the next stage. In total, the truck travelled 4,892.5km during the 2016 race, spending 80 hours on the road. 
In 2015, the data was made available to commentators, the media and fans via a beta version of a live tracking website, which was developed to support 17 million views and up to 2,000 page requests per second, while also being optimised for Apple, Android and other mobile devices.

The website, coupled with the live data feeds over social media, enabled ASO to provide a richer and more relevant viewing experience to modern cycling viewers and fans. One important additional consideration around the live data feeds is cyber security. Hosting in the Dimension Data cloud ensures the data is kept secure, and encrypted connectivity allows the race information to be transmitted to broadcasters, the media, the teams and viewers in a secure way.

Dimension Data’s Big Data Truck – a fully mobile data centre – located in the technical zone near the finish line of each stage
Dimension Data’s Big Data Truck – a fully mobile data centre – located in the technical zone near the finish line of each stage

During the 2015 race, the solution was enhanced many times. To ensure optimal performance, Dimension Data’s team on the ground collaborated with support teams situated in Australia, India and the US, using live video, chat feeds and other collaborative tools to continuously enhance and improve the platform. Having teams in multiple time zones enabled Dimension Data to take a round-the-clock development and testing approach to adapt the solution during and after each stage throughout the month-long Tour.

With the eyes of the world on the race, it had to ensure the solution was both secure and fail-proof, with 100 per cent uptime during the race stages. This was no mean feat in remote environments, such as mountain top finishes or Cols, or during times of bad weather or frenetic rider activity, including crashes in the peloton.

The Dimension Data solution enabled ASO to ramp up new environments rapidly, scale technology platforms up or down on demand, and respond to changes or new requirements almost immediately. ASO wasn’t required to make large capex investments in depreciating technology assets; Dimension Data’s cloud based solution was simply ‘put into hibernation’ until the next cycling race.

Enhancements for the 2016 Tour

A number of significant improvements have been made to the solution since the 2015 Tour de France. For example, in 2016, a VCE VxRail enabled virtual SAN solution was added into the mobile data centre for redundancy. In addition to being easy to use and set up, VxRails take up far less space, meaning that the solution can be operated with only half a rack, making hardware more stable and removing risk in the stack when in transit.

The upgraded solution has reduced complexity in the network and in its compute and storage, with the aim of being more operationally efficient. The set up and shutdown process at the start and end of each stage was also reduced by at least 20 minutes each day, sometimes saving as much as an hour a day.

In 2016, the transmission range of the solution was also enhanced, giving a 10-fold increase over the previous year. This meant fewer dropouts or ‘gaps’ in race data, seamless communication and continuity throughout the race.

Dimension Data’s Big Data Truck was equipped and set up so that it could accommodate various television graphics teams, race coordination officials, data capturers and analytics teams, responsible for delivering the complete end-to-end data solution at the 2016 Tour de France. This is one of the biggest changes from 2015, where each team worked separately, and continues the theme of collaboration and data integration made possible by the advancements in the technologies on display.

Also new for 2016 was a web based application called ‘Race Center’, which combined the live race data, video, photographs, social media feeds and race commentary, becoming the official digital hub of the race. The new live tracking website gave viewers an immersive digital experience with access to more accurate information from riders in the 22 teams, including speed, distance between riders, composition of the pelotons, wind speed and direction, as well as prevailing weather conditions.

Chris Froome, winner of the 2016 Tour, was forced to run part of the race following a crash with a motorbike

“The enhancements to this year’s solution means we were able to tell richer and more enhanced stories as they happen, giving viewers, the media, cycling fans and race commentators deeper insights into the sport that weren’t available until now,” says Peter Gray, Dimension Data’s sports practice senior director of technology.

“This year, we worked with a much broader palette, which means access to more meaningful race data, race routes, riders and current weather conditions. What’s been exciting is the ability to deliver all this information to ASO through a unified digital platform. This makes the quality of the data even more valuable for viewer engagement, and speaks directly to a generation of younger viewers who rely on new technologies such as social media and live video to engage with their world.”

Christian Prudhomme, director of the Tour de France, ASO, says, “The unprecedented growth in different social channels like Instagram, Twitter, Facebook and live video at last year’s race, demanded these technologies be embraced and enhanced for modern viewers. The Tour de France is a flagship event in a modern world, and it’s only natural that we give our viewers access to as much quality content, entertainment and analysis as possible through the media they use every day. Together with Dimension Data, we’ve been working on new ways to appeal to our billions of viewers, and we’re excited to showcase the result of our efforts through Race Centre.”

Christian adds, “I believe the appeal of having access to multiple real time video, social media and live race information from one responsive and intuitive interface, greatly enhanced the quality of coverage of the Tour de France, and will become an essential companion to the largest live televised event in the world.”

As technology continues to evolve, the possibilities for capturing data in professional cycling, and sport in general, are expanding. Being able to send large amounts of data from a large group of riders is already enabling greater analysis, from learning where the stresses and strains are on a bike, to how a rider’s position on their bike can improve performance. As more data is collected, this will fundamentally change sport, from how riders train to how viewers, fans, and the global media consume live events.