Aston Martin F1
The Aston Martin Formula 1 team (AMF1) is midway through a five-year plan to get to the top of the sport. Now in year three, it gets regular podium finishes with four in the past 12 months and its AMR23 car is its fastest yet.
Use of trackside compute and the storage it needs is vital in F1 to track and adjust car componentry to get the kind of gains seen by the AMF1 team.
Core to achieving that has been a partnership with NetApp that the storage vendor is keen to showcase as an example of performance in its NVMe-equipped hardware, but also its ability to manage data efficiently across remote locations, such as via replication and handling ageing data.
Since coming on board with AMF1, NetApp’s contribution has been to receive, retain and transmit race data for analysis in real-time and post-event for a team of 60-plus people dedicated to analysis and engineering changes to vehicles.
“The cars have around 300 sensors that transmit data constantly to relay temperatures, wear, speeds, etc,” said Aston Martin F1 CIO, Clare Lansley. “These allow us to see how close to what we thought performance parameters would be and make changes. Then raceday data is tagged with metadata that includes weather, time, driver details, and so on.”
The hardware used trackside comprises of two redundant FlexPod converged compute and storage stacks. Storage is made up of two NetApp A250 HA pairs with 3.8TB of NVMe capacity in internal bays with 10Gbps connectivity. Compute is Cisco with 168 cores and 2TB of memory.
Race event analysis goes into helping make the cars perform better by being able to make marginal gains across multiple car components. Being able to do it all efficiently in IT terms enables Aston Martin to put more resources into the physical aspects of racing.
So, key decisions for the Aston Martin team exist around how long to keep data for – or, more precisely, how long it should be retained on the most performant tiers of storage.
“FIA regulations change weekly, so we constantly have to make decisions about how much of the data we keep is worthwhile. And it’s about costs,” said Lansley. “There are cost caps imposed by the FIA. And where there are IT efficiency gains, it means I can give more time back to focus on the car.”
“Form factor is also key. Not only do we need the performance, density and redundancy, but it all has to fit into the ‘freight jigsaw’ we use to move everything to an event.”
Efficiencies include use of NetApp SnapMirror replication which has cut data transfer windows from remote locations in half. Previously, data transfer had been by manual copying.
Lansley also said they saved weight compared to the previous setup and have had no outages since the Flexpods were deployed.
NetApp provides the storage at Aston Martin’s datacentre too, with an A400 with 15.3TB of flash storage on board.
NetApp and Aston Martin’s initial engagement had been just for trackside hardware. NetApp approached AMF1 with proposal’s to revamp its tech, which had been based on a mix of vendor technologies, that were described variously as “complex” and “legacy”.
Stage two saw NetApp propose “an Aston Martin racing fabric” with larger FlexPod deployments at AMF1 team HQ and connectivity via SnapMirror.
Stage three, it is hoped, will see extension of the existing infrastructure to the cloud, which could perhaps receive cold data for long-term storage where no egress is likely; for backups, likewise with limited likelihood of retrieval; and as an experimental area to develop and test new applications.
“Cloud will give us the ability to spin up test environments quickly [and do] agile testing,” said Lansley.
Read more about tech in F1
- CIO interview: Clare Lansley, CIO, Aston Martin Formula One. A rare female CIO in a male-dominated sport, Lansley discusses how digital transformation is all a part of helping the team to move further up the F1 grid.
- CIO interview: Michael Taylor, Mercedes-AMG Petronas Formula One. We speak to the IT director of the Mercedes-AMG Petronas Formula One team about historical data, race data and fast-paced analytics.