SailGP charts a new era of competitive sailing with Oracle Cloud Infrastructure

February 8, 2022 | 5 minute read
Natalie Gagliordi
Senior Writer
Text Size 100%:

Courtesy of SailGP

When Oracle Founder and CTO Larry Ellison and champion yachtsman Russell Coutts envisioned the Sail Grand Prix racing league, they started with a blank piece of paper—for the competition, and for the technology architecture that supports it.

“We had no legacy software, no legacy hardware,” says Warren Jones, chief technology officer at SailGP. “It was, how can we future-proof this for the next 5, 10, 15 years?”

Today, SailGP is a high tech global racing series in its second season, with eight national teams competing on identical F50 foiling catamarans capable of reaching speeds of up to 60 mph. Both the boats and the athletes who sail them have sensors that collect and stream reams of data for every competition. And the technical infrastructure behind that has come a long way from that initial blank slate. 

For its inaugural season in 2019, SailGP relied on on-premises Oracle Exadata at the league’s core locations in San Francisco and Bermuda to collect and manage data. The competitions are now held in destinations all over the world, and SailGP has moved away from physical infrastructure and into Oracle Cloud Infrastructure (OCI).

“With SailGP moving around the world, we decided that the cloud infrastructure was going to be the key for what we wanted to do,” Jones says.

SailGP relies on Oracle Autonomous Data Warehouse to handle all that race-related data, with streaming analytics, machine learning, and visualization technologies all running on Oracle Cloud Infrastructure without database administration. In a matter of seconds, SailGP analyzes more than 240,000 data points collected from 800 sensors on the F50 vessels, streaming racing metrics in real-time to crews.

SailGP crews analyze key metrics during a race such as speed and rudder differentials to improve the performance of the team. Some of the F50s collect simple data from pressure sensors, while others collect more complex data from gyroscopes and GPS. Sensor data flows off the boats into Oracle Autonomous Data Warehouse in real time. “The average batch size is 15,000 to 20,000 messages, every 500 milliseconds,” says Aleksandar Kocic, SailGP senior solutions architect, during an episode of Built & Deployed, a video series featuring software architects discussing cloud technology.

The data points collected from SailGP’s catamarans—which include boat altitude, speed, pitch, water, and wind conditions, as well as sailor biometrics—are transferred from the F50 vessels to an Oracle Autonomous Data Warehouse using a dedicated 1Gbps Oracle FastConnect link. This database provides each team with access to real-time and historical data, Kocic says. 

The data from the F50s are distributed to compute instances for preprocessing in the media data systems server and PI server. That data is then transformed from a proprietary format to JSON and then sent to Oracle Stream Analytics, which processes the data and detects relevant events.

The data is then sent to Oracle Autonomous Data Warehouse. “The primary goal of the ADW is to achieve fast, complex analytics for further data manipulation,” Kocic says. Oracle Autonomous Data Warehouse is also used to make data available to media partners through Oracle REST Data Services and to get data from SailGP’s MySQL Database Service on OCI, which is mainly used for legacy applications. The data from MySQL Database Service is replicated to the data warehouse using Oracle OCI Golden Gate.

Jones is most excited for what’s next as the team taps the capabilities built into Oracle Autonomous Data Warehouse.

“To have machine learning and AI associated with SailGP is truly amazing,” Jones says. “We've got so much data available to us, we haven't touched the ability of what we can do at the moment. In the future, we know that there's going to be insights that we haven't really thought of today, which will be more relevant in two, three, four years’ time.”

Anticipating mechanical failures

Data from the F50s also help teams detect anomalies and proactively manage the boat’s moving parts should an issue with a particular component arise before, during, or after a race. This not only improves performance, Jones says, but also helps the teams save time and money.

“If a part breaks on the boat, then that appendage doesn't just break, it may break other things around it,” Jones says. “So, by replacing the part when it is no longer fit to do its job, we save time on a lot of other things as well. We think in the next year or so, we're going to get some hard data to give us quantifiable timings of appendages on the boat, and really utilize that data to help fix and take away parts.”

In the SailGP league, all technology and performance data are shared with broadcast partners, fans, race officials, and importantly, across the fleet—each team has full access to one another’s data for post-race analysis. The idea is that sailing performance on the water, not technology, should determine the winner each race. 

SailGP’s open data policy is unique for a competitive sports organization. In most sports, the data generated by a team is proprietary and closely guarded. SailGP is also using OCI-powered data and analytics to assess and lower the carbon footprint of the league’s events.

“I think this is a fundamental shift in what sports do in using data to go faster or be more efficient,” Jones says. “We're really trying to push the boundary of technology, sustainability, and the use of tech for good. And with Oracle's help, we're really pushing the boundaries.”

SailGP also uses data generated from the F50s to serve the SailGP mobile app, SailGP Insights powered by Oracle, and Liveline, an augmented reality graphics package that gives racing fans more insight into happening on the course.

“In the future, we want to really push this, and give fans the ability to choose what they want to see and how they want to see it,” says Jones.

Natalie Gagliordi

Senior Writer

Natalie Gagliordi is a senior writer at Oracle. She spent a decade as a technology journalist and was previously a senior writer on ZDNet.

Previous Post

Enhancing Database and Apps Troubleshooting with ML Driven Log Analytics using EM OCI Bridge

Kumar Varun | 4 min read

Next Post

Modernize your motor sport race strategy on Oracle Cloud Infrastructure

Arnaud Froidmont | 3 min read