“ For our wants, AWS outperforms all different cloud suppliers, in velocity, scalability, reliability, international attain, companion neighborhood, and breadth and depth of cloud companies obtainable.”

Pete Samara
Director of Innovation and Digital Expertise, Formula 1

Formula One (or F1) has been one of many premier types of auto racing all over the world because it’s inaugural season in 1950. Formula One vehicles are the quickest regulated road-course racing vehicles on this planet, owing to very excessive cornering speeds achieved by the era of huge quantities of aerodynamic downforce.

Formula One is not only a battle between the world’s greatest drivers, but in addition a battle between a number of the world’s most modern engineers. No different sport has been as dynamic in its evolution and embrace of recent expertise. Whereas a number of the tech goes to serving to drivers, who’re hitting speeds as excessive as 230 MPH, taking pit stops in below 2 seconds, and flying round corners with a pressure of 5G, a lot of it goes to reinforce the experiences of its rising base of over a half a billion followers.

{Data} in Formula One

An F1 race automobile, with a mean of 150–300 sensors on board, generates greater than 1.1M {data} factors per second, transmitted from the vehicles to the pit. Formula 1 is a really data-driven sport the place a lot of the fun comes from extracting thrilling particulars on efficiency statistics. F1 depends on the breadth and depth of AWS companies to stream, course of, and analyze that flood of information in real-time, after which current it in a significant approach for F1 international TV viewers.

For such excessive efficiency battles of velocity and management, an unlimited quantity of computation, info gathering, analytics, and studying is required, earlier than, throughout and after the races. So in 2018, Formula One determined to maneuver the overwhelming majority of its infrastructure from on-premises {data} facilities to AWS, and standardizing on AWS’s machine studying and {data} analytics companies to speed up its cloud transformation.

Amazon Web Services (AWS) is the official cloud service and machine studying supplier for Formula One.

To reinforce race methods, {data} monitoring methods, and digital broadcasts, F1 is working with AWS with quite a lot of companies akin to Amazon SageMaker (Platform to deploy machine-learning fashions), AWS Lambda (event-driven server-less computing platform), and AWS analytics companies — to uncover a broader perspective of the metrics concerned in racing. AWS additionally performs a significant position in Computational Fluid Dynamics (CFD).

Computational Fluid Dynamics

Formula One has been utilizing Amazon EC2 for Computational Fluid Dynamics (CFD) to simulate race automobile aerodynamics, attaining the efficiency of an excellent pc at a a lot decrease price and decreasing simulation time by a mean of 70 % — from 60 hours right down to 18 hours. with the CFD mission, Formula One used over 500 million {data} factors to check downforce loss when two automobiles race in shut proximity (downforce will increase tire grip and cornering speeds and reduces lap time). Primarily based on its CFD simulations, Formula 1 has designed a automobile for the 2021 racing season that reduces downforce loss in wheel-to-wheel racing from 50 % to fifteen %.
Computational Fluid Dynamics Undertaking makes use of over 12,000 hours of compute time to design the automobile for the 2021 racing season.

CFD simulates the impression of a liquid or gasoline on an object and requires in depth compute capability to carry out this type of simulation, requiring excessive efficiency computing (HPC) clusters to do the job.

To finish the CFD work, Formula 1 used AWS ParallelCluster on Amazon EC2 to run the OpenFOAM CFD framework, and Amazon Easy Storage Service (Amazon S3) for {data} storage. Leveraging the scalability of the cloud, Formula 1 was in a position to run CFD simulations on core counts a lot bigger than they had been beforehand in a position to execute. The elevated velocity with which the aerodynamics crew may run detailed, two automobile turbulence simulations on AWS meant they may enhance the variety of automobile designs they may examine from one to 5 per week. Transferring ahead, there are plans to broaden the appliance additional, as much as 2,300 cores, and to introduce AWS Machine Studying (ML) instruments, akin to Amazon SageMaker, to permit ML applied sciences to assist with the design and additional optimize the efficiency of the automobile.

“This mission with AWS was probably the most revolutionary within the historical past of Formula 1 aerodynamics, … We now have been in a position to make use of AWS applied sciences to grasp the unbelievable aerodynamic complexities related to multi-car simulations, and are happy that the outcomes point out we’ve got made wonderful progress in direction of our goals of nearer racing.”

Pat Symonds
Chief Technical Officer of Formula 1

Different Companies

• Utilizing Amazon SageMaker, Formula 1’s {data} scientists are coaching deep studying fashions with greater than 65 years of historic race {data}, saved in each Amazon DynamoDB and Amazon Glacier. With this info, Formula 1 can extract essential race efficiency statistics to make race predictions and provides followers perception into the split-second selections and techniques adopted by groups and drivers.
• Formula One additionally chosen AWS Elemental Media Services to energy its video asset workflows, enhancing the viewing expertise for its 500 million plus followers and viewers worldwide.

• By streaming real-time race {data} to AWS utilizing Amazon Kinesis– a service for real-time {data} assortment, processing and evaluation, Formula 1 can seize and course of key efficiency {data} for every automobile throughout each twist and switch of the Formula 1 circuits with unmatched accuracy and velocity.
• To create new insights, Formula One makes use of 70 years of historic race {data} saved in Amazon Simple Storage Service (Amazon S3), mixed with dwell {data} that’s streamed from sensors on F1 race vehicles and the trackside to the cloud by Amazon Kinesis.

The Automotive Efficiency Scores (makes use of a spread of AWS companies) perception shows an on-screen graphic that gives followers with an entire breakdown of a automobile’s complete efficiency utilizing 4 core metrics: Low-Velocity Cornering, Excessive-Velocity Cornering, Straight Line, and Automotive Dealing with. The brand new graphic illustrates how these metrics examine from one automobile to a different, enabling race followers to gauge a given automobile’s relative efficiency in these totally different areas and see the place every crew and driver is main the pack or shedding essential time to their rivals.

Tyre Efficiency graphic (powered by AWS) gives real-time info on the present situation for all 4 tyres on a selected automobile, introduced as a share worth. The dimensions runs between ‘new tyre with no put on’ (100%) and a ‘used tyre’ on the finish of its efficient efficiency lifespan (0%).

F1 is ready to analyze race efficiency metrics in real-time by deploying these ML fashions on AWS Lambda, which runs code with out the necessity to provision or handle servers. All the insights will probably be built-in into the worldwide broadcast feed of F1 races across the globe, together with its digital platform F1.television, serving to followers to grasp the split-second selections and race methods made by drivers or crew strategists that may dramatically have an effect on a race consequence.

Formula One has collaborated with AWS for a number of the greatest and correct statistical {data} within the motorsport world. To mix excessive velocity physics, {data} analytics, efficiency, and lots of different components, right into a packaged service has allowed Formula One to current a extra detailed and thrilling viewpoint of the game to its viewers.

“Formula 1’s years of helpful historic race {data} analyzed towards the real-time info that’s collected in each race utilizing AWS’s machine studying, streaming, and analytics companies will uncover new racing metrics and insights that had been unimaginable previously …
Formula 1 racing mixes physics and human efficiency, yielding highly effective, however advanced {data} that AWS helps them to harness.”

Mike Clayville
Vice President, Worldwide Business Gross sales at AWS