How big data is helping to power Mercedes to F1 glory

Flash storage and advanced analytics allows Mercedes-AMG Petronas Motorsport to hone performance in simulations and out on the track.
Written by Mark Samuels, Contributor

Mercedes-AMG Petronas Motorsport has created a big competitive advantage in a sport of fine margins. The F1 racing team, home to recently crowned world champion Lewis Hamilton, is celebrating its fifth successive constructors' title.

That level of success gives Matt Harris, head of IT for Mercedes-AMG Petronas Motorsport, an opportunity to reflect on the part his technology organisation has played in the team's winning journey. Harris has been with the team in various guises for two decades, becoming head of IT in August 2009.

"The amount of time I've been with the company probably tells you how great this job is -- the big thing for me is it's always a challenge and it's always different," he says. "We're pushing the boundaries of technology and trying to get the best out of IT. It's the key way we can differentiate ourselves from other teams."

Harris, who spoke to ZDNet at the Big Data LDN event in London, has seen tremendous change at his company and in the sport during his time in F1. He says the team is more like a business these days, albeit one with a very public-facing front. One of the biggest changes during the past five years has been the growing reliance on technology and data.

"IT is integral to everything we do, whether that's for the main car, the race team or the supporting functions," he says. "We have to make sure we're at the top of our game all the time. We can cause the car to stop very easily if IT fails."

When it comes to positively affecting performance, Harris says the impact of the IT department is felt in a more indirect form. The aim is to provide systems that allow people on the team to make accurate decisions quickly.

SEE: 60 ways to get the most value from your big data initiatives (free PDF)

The team races two cars a year and the unique conditions of each Grand Prix circuit means the cars are modified for each race. Those modifications take place in an iterative manner -- and technology and data are crucial to these changes across all areas of operation, from design to testing and onto racing.

Each car produces half a terabyte of data during the average race weekend, with a further 5 to 10 TB of data produced per week at the team's headquarters in Brackley, UK. With the margin of victory being measured in thousandths of a second, rapid processing and analysis of data is key to victory.

"Data's critical -- without it, we can make very few decisions," says Harris. "That data can be both structured and unstructured. Just because a driver turns around and tells us something, we can't take it for granted -- we prove it with data. We look for the anomalies in data that support configuration changes on the car."

He says simulations prior to race day are crucial. The team will make multiple iterations during this simulation process. The faster the team can do this work, the greater the impact -- and reliable IT is critical to this process. Harris says the IT department has made big progress in this area during the current F1 season.

"It used to take us three days to run simulations and process the results. That realistically meant you had one chance to run a simulation during a week between race days. We've now reduced that process to one day -- so you can run a simulation on Monday, look at the results, seek improvements, and run further simulations on a daily basis until race day."

The team is involved in a constant set of changes and IT must support this. Harris says one of the key areas of transformation this season has centred on both access to data and effective visualisation. Harris says IT partnerships with Pure Storage and Tibco have been crucial.

Pure has helped Mercedes F1 remove storage bottlenecks associated to older spinning-disk hardware. In the past, Harris says one of his staff members had to spend three or four days a week looking after this older kit. Now with flash, the management process is automated, and the only requirement is to ensure the team is making optimal use of its storage.

Since buying its first FlashArray from Pure in 2015, the team has added several more. It has now moved virtually all its production applications and data onto Pure arrays. Harris says flash storage provides the foundations for effective data management at the track and back in the factory.

"We're putting the kit through a very harsh environment. We're expecting it to work at optimal performance. It needs be reliable, always on, and we're always creating more data," he says.

SEE: Sensor'd enterprise: IoT, ML, and big data (ZDNet special report) | Download the report as a PDF (TechRepublic)

Tibco's big data tools run on top of the Pure storage foundations. Harris says these tools allow the team to note anomalies in performance data. The team uses the Tibco Insight Platform for advanced analytics. The platform provides visual, predictive, and real-time data streaming relating to car performance.

"Technology helps us to identify the unexpected and to make changes to the car that create the right results," he says. "We're solving sporting problems, rather than simply trawling through repetitive and mundane information. As an IT team, we have to make sure that whoever makes the decision has the data all the time at their fingertips."

Harris says the change in focus from monitoring to analysing helps his people feel as if their contribution makes a significant difference. "We're making sure that -- from the top to the bottom -- the company's and the employees' objectives are aligned," he says.

"People in the IT organisation now feel engaged and part of a team, not just a support department in the corner that's a cost. Making sure everyone recognises their job is important has been a key consideration for us in the past couple of years."

That focus on teamwork is as true for the team's racing drivers, Lewis Hamilton and Valtteri Bottas, as it is for the IT professionals. Harris says the combination of teamwork, capability and technology means the team is in pole position for further racing success.

"The good thing with the drivers is they come around the factory once a year at a minimum and they'll spend an hour in the IT department talking with everyone," he says. "They want to understand what we're doing -- at the end of the day, they're just another employee. Yes, they have an important job during race weekend, but so does everyone else."


AMD-powered Cray to speed up Formula One racing

The Haas F1 team is leveraging a Cray CS500 cluster supercomputer powered by AMD EPYC 7000 processors to carry out computational fluid dynamics simulations.

Formula 1: How sensor technology is changing the race

With Formula 1 pushing its own digitisation projects, each team has also formed a series of close technology partnerships such as that between AT&T and Aston Martin Red Bull Racing.

NFL in London: How big data is helping to change the game when it comes to player performance

New technologies are giving NFL coaches, and football fans, better insights into how players are performing.

How investing in IT helps to deliver the goods

Tech is a key differentiator for Hermes, which is looking at cloud, big data and future innovation, says the logistics company's CIO Chris Ashworth.

AI and machine learning: Building use cases and creating real-life benefits

One organisation, two very different uses of AI and machine learning.

AI, cybersecurity shape the CIO agenda for 2019 as IT budgets rise (TechRepublic)

Companies are scaling digital transformation projects, but privacy remains a top concern, according to a Gartner report.

Toyota and SoftBank are teaming up to bring big data to mobility (CNET)

The two companies are forming a joint venture called Monet Technologies.

Editorial standards