F1 telemetry: The science of speed data-ing

There’s a data war going on in Formula 1 and, although we hear about it only sporadically, it’s central to the success – and failure – of every team.

For the first 25-30 years of F1, there were almost no computers to help engineers and aerodynamicists build better cars.

Slide rules and log tables, pencils and protractors were the tools of choice, and it’s fair to say the racing was no less spectacular because of it.

While we now take computers and data crunching for granted in F1, they didn’t really make their presence felt properly until the 1990s – in the years before that, even an electronic fuel readout was considered supremely high-tech…

Telemetry: Talk to me

Electronics started to appear in a very limited way in the 1960s but it would be years before the earliest electronic breakthroughs in F1, which revolved around basic telemetry and equally basic engine management systems.

McLaren began experimenting with telemetry – downloading basic data from a car – in 1975, but that was in IndyCar racing.

A grand total of 14 sets of information about the car could be downloaded when the car came to a halt. These days, the smartphone in your pocket is capable of monitoring its environment more thoroughly than a 1970s F1 car was.

The pace of change accelerated in the 1980s, as computing power began to grow rapidly.

Engine management systems combined control of ignition and fuel injection in a single unit, giving teams new-found control over power, efficiency and driveability.

Hard though it is to believe now, something as simple as electronic fuel readouts were a major breakthrough in the middle of the 80s, helping drivers deal with fuel restrictions and, on plenty of occasions, lulling them into a false sense of security with inaccurate readings too.

Rise of the robo-cars

The 90s were the decade in which electronics utterly transformed the sport. Indeed, such was the pace of innovation that drivers were in danger of being eclipsed by the abilities of their digital assistants.

Active suspension controlled cars’ ride heights and helped aerodynamics no end; electronically assisted steering took some pressure off drivers in the corners; braking aids helped cars slow down and traction control helped them speed back up again without lighting up the rear tyres.

While many driver aids would be banned, there was a lasting legacy from this period. Huge amounts of data were being captured from cars, transmitted to team garages in bursts of activity as cars passed the pits.

This ‘burst telemetry’ was introduced in the late 80s and really came into its own in the 90s. Engineers in the pits and back at team HQ could analyse information about the car before it had even stopped.

Nowadays, F1 teams use the internet to stream vast amounts of data from circuits to their HQs, with connections that are at least ten times faster than your domestic broadband.

It’s hard to over-estimate just how much difference this has made to the sport.

According to McLaren software engineer Chris Alexander, sending a three-minute music track would take hundredths of a second on the team’s internet connections.

By contrast, the first real digital network – ARPANET, in the US in 1969 – would have taken more than five hours to perform the same task.

Remote control

The transmission of information is not one-way. While data from cars is streamed in real time from circuits to engineers at a team’s HQ, these engineers are also sending information back to the track.

Why not just have the entire engineering team trackside anyway? In a word, cost.

It’s just not practical to haul everyone and their kit around the world, although engineers at a circuit will have the best laptops to access all the data flying around.

Back at base, super-powerful workstations help strategists and engineers analyse the information, running specialist software such as SAP HANA to search through and analyse the vast amounts of data at their disposal in real time.

Inevitably, in the innovative world of F1, teams don’t just rely on third-party solutions.

McLaren, for example, has developed its own sophisticated platform for analysing data and running simulations.

It lets every engineer on the team, both at the Woking HQ (pictured) and at circuits, access massive amounts of information.

McLaren’s Chris Alexander said: ‘This platform unifies access to a wide variety of data – from the cars on track at grands prix to laps driven by our test drivers in the simulators, and from aerodynamic data generated by the windtunnel to specialised test equipment for specific car components such as the clutch or brakes.

‘Because all of this data can be accessed in the same way, new exploratory and analysis tools can be quickly and easily developed for emerging needs, as and when they occur in the fast-paced environment of Formula 1.’

Despite this unified approach to data, different specialists within a team will also have their own dedicated software packages to help them get to grips with whatever information is of most importance to them.

Chassis specialists, for example, will be running different analytical packages to brake experts.

Up in the clouds

While F1 is an immensely innovative environment, it’s not slow to follow the lead of others when new approaches become available.

So, not surprisingly, all teams make use of cloud-based computing, calling on virtual computers in data centres around the world to help them number crunch quickly in a way that simply could not be matched by the team’s own computers.

That’s not to say that every piece of data is floating around in the cloud.

Teams need sensitive information to be both available to them and secure from prying eyes.

So custom solutions are also utilised. For example, earlier this year Mercedes teamed up with Pure Storage as they sought to develop a data infrastructure that was powerful and also portable.

Pure Storage said: ‘Putting performance-critical workloads at risk of availability in the cloud was a risk the team could not afford.’

At the heart of the system are two FlashArray//m70 arrays and four FlashArray//m20 arrays in three pairs.

To give you an idea of the computing scale this represents, a single FlashArray//m70 array offers 1.5petabytes (PB) of data storage – that’s 1.5million gigabytes (GB). It rather puts the 256GB in your shiny MacBook Pro to shame.

And what does 1.5PB look like? It’s equivalent to 750billion pages of standard printed text. Yes, billion.

Of course, computers crop up in countless other areas of F1, not least in the offices of aerodynamicists, but that’s a topic in itself and a subject for another day.

The pace of digital innovation is as bewildering in F1 as anywhere else, and the modern F1 world would be incomprehensible to those who were working in the sport in the first half of its history.

It’s hard to imagine what an engineer with a slide rule and log tables would make of the idea of those 750billion pages of electronic storage – but that’s the sort of change that relentless innovation can bring.