AWS drives cloud and machine learning for Formula 1

AWS drives cloud and machine learning for Formula 1

The Formula One Group has announced that Amazon Web Services (AWS) will be hosting the majority of its IT infrastructure, in a wholesale shift into the cloud and away from on-premise data centres.

Formula 1 will also employ AWS’s machine learning and data analytics services to provide insights into race strategies, and help it with data tracking systems and digital broadcasts.

One of the tools at the centre of the new strategy is Amazon SageMaker, a machine learning service that enables developers to build and deploy machine learning models easily.

SageMaker is helping F1’s data scientists to train deep learning models with over 65 years’ worth of historical race data (stored in Amazon DynamoDB and Amazon Glacier). The hope is that they will be able to extract valuable performance statistics to make predictions and offer fans insights into racing teams’ competing strategies.

Other critical uses of the technology will be on the track itself, and include predicting the best times to pit cars and change tires for maximum advantage.

Pete Samara, director of Innovation and Digital Technology at Formula 1, said:

By leveraging Amazon SageMaker and AWS’s machine learning services, we are now able to deliver these powerful insights and predictions to fans in real time. We are also excited that the Formula 1 Motorsports division will run high-performance compute workloads in a scalable environment on AWS.

“This will significantly increase the number and quality of the simulations that our aerodynamics team can run, as we work to develop the new car design rules for Formula 1,” he added.

Beyond this, AWS Lambda – AWS’s event-driven, server-less computing service – and its analytics services will allow Formula 1 to tap into new metrics “that will change the way fans and teams enjoy, experience, and participate in racing”, said the F1 Group. Meanwhile, AWS Elemental Media Services will underpin its video asset workflows.

Formula 1 also plans to use Amazon Kinesis to capture real-time performance data, which can be analysed to assess driver performance and then relayed to fans watching on TV or online, giving them new insights into the sport and the inner workings and decisions of each team and driver.

Mike Clayville, VP of Worldwide Commercial Sales at AWS, said:

Formula 1’s years of valuable historical race data, analysed against the real-time information that is collected in every race using AWS’s machine learning, streaming, and analytics services, will uncover new racing metrics and insights that were unimaginable in the past.

Internet of Business says

Last week, Internet of Business reported how IBM Watson’s partnership with the Wimbledon tennis championships is providing AI-assisted video editing, based on recognising players’ and fans’ emotions. Now another major tech player has seized the opportunity to leverage the popular appeal and profile of a sport to champion its products.

AWS’s suite of cloud tools boasts the range of applications that Formula 1 needs to peel away the opacity of motorsport for anyone who isn’t already a devoted fan. This will allow the average viewer to engage with races and better understand the margins that can mean the difference between poll position and middle of the pack – all in near real time.

The top-secret technical war between Formula 1 teams, that, season after season, is characterised by minor aerodynamic changes, engine upgrades, or regulatory changes, is not dissimilar to the marginal struggles for increased efficiency that occur between competing businesses.

This is why F1 offers an ideal proving ground for AWS, and a global stage on which to showcase its cloud tools, to an audience that includes influential business leaders and high-profile sponsors.