McLaren CIO Stuart Birrell has said that he is using SAP’s in-memory HANA technology to try and scale back on using “expensive PhD” skills within the group’s Formula 1 teams.
Speaking at an SAP Forum event in London this week, Birrell told audiences that data is fundamental to the racing team’s success.
“We generate gigabits of data every weekend off every car and analyse that in real-time to make decisions,” said Birrell.
“We monitor everything - the movement on the suspension strap, what’s happening in the engine. There are probably 120 sensors in each Formula 1 car.”
The information generated by the cars during a race is sent back to McLaren’s on-site garage in real time and then back to the head office in Woking.
He added: “Getting that information is critical. We probably run one thousand simulations a minute during a race, so by the end of the first or second lap we have a 90 percent probability of the outcome.”
McLaren also uses the data to retrospectively analyse its cars’ performance on specific tracks to help the driver understand how to best race the car.
“Monaco’s track hasn’t changed for 30 years, so we can look at years’ worth of data and see what happens to a suspension arm on certain corners and monitor that,” said Birrell.
Birrell stated that the company’s investment in data over the last 10 years had been “huge” and this is why it was now looking to SAP’s HANA.
“We are working with SAP on the HANA platform to turbo charge our analysis. It needs that next generation,” he said.
“It needs stepping up a gear to get it to that last one percent. Many races are often won within seconds and finding that small percentage difference between us and Red Bull or Mercedes is absolutely fundamental.”
HANA uses in-memory software to analyse huge volumes of data in real time. SAP began looking at in-memory capabilities four years ago and started developing the technology in early 2010.
A proof of concept was established in October of 2010, which saw SAP's data analytics improve by a factor of 14,000 - a request now takes one second where it previously took five hours.
The increase in speed is gained by putting data analytics into a CPU, into 'memory', as opposed to being carried out on disk.
Queries sent to on-disk databases take longer because of the mechanical interaction that has to take place, which is removed when being conducted in-memory.
Birrell said that he is ultimately looking to HANA to help cut costs on expensive data scientists that currently help with the team’s data analysts.
“Using HANA we can hopefully automate decision making. People have always made decisions based on the data, but we want to get to the point where the system can make the decision,” he said.
“We would then need less PhDs (postgrads) in the garage. PhDs are very expensive.”
It was also revealed yesterday that a UK government department is looking to use HANA to help consolidate applications and cut further costs.