Speedboats race with the cloud
SilverHook Powerboats uses a private cloud to analyze and visualize racing data in real-time -- at 140 miles an hour
By Sharon Gaudin
Computerworld | Oct 28, 2014 2:54 AM PT
The founder of a powerboat racing business is looking to use the cloud to track the company boat's performance during offshore races and keep fans and racing officials updated with animated race visualizations.
http://youtu.be/KXeb9MTWOug
"It really pushes the envelope..., being able to develop [a cloud technology] quickly and having a robustness of coverage during an event," said Nigel Hook, co-founder and president of San Diego-based SilverHook Powerboats. "We want to do things that would not be possible without the cloud."
SilverHook Powerboats sells and races high-performance boats. Starting Nov. 5, it will run its first test of its new cloud system during the 34th Annual Key West World Championships. A prototype will run live during the races scheduled for Nov. 5, 7 and 9.
The system, which will only be working for the one SilverHook boat, is set up to transmit real-time information, ranging from GPS positioning to engine performance data, oil pressure, fuel pressure, horsepower and acceleration metrics. That data will be collected and analyzed in a private cloud that IBM helped SilverHook set up so the team's crew chief can access the information during the race and give the crew better instructions.
If all goes as planned, many, if not all, the boats in a race will at some point transmit data to the cloud -- and all of the racing teams, race officials and fans will have access to information about boat performance, speeds and even ocean conditions.
"We're out in the ocean in this brutal environment," Hook said in an interview. " "We're up to 6.9 Gs in a race. There's the salt water. The wind. You're flying in six-foot seas. What makes it tough for the people in the boat is when you're taking that kind of impact, it's very tough to read the gauges in front of you. That's where it's very important for us to have all the data transmitted up into the cloud so the crew chief can see if there's a problem, like with an engine, that we can respond to."
And in a sport that takes place sometimes miles offshore, real-time animations of the race would let fans feel closer to what's happening and follow their favorite boat or driver.
"If the other boats in the race were doing the same, then all of their data and our data is in the cloud and then you have a rich set of data to share with the fans," explained Hook. "That brings them much closer into what is going on. Even if a camera is not on you, a fan could visualize the data coming out based on speed and GPS position and G forces. Those are just numbers, but if we could take that information and show an animated picture on the computer, a fan could see what's happening without just seeing a scroll of numbers."
He also hopes to one day equip boat drivers with wearables to monitor and transmit their blood pressure, temperature, heart rate and oxidization levels, as well as the boat's stats. "...That's the next level," said Hook. "You see what's happening to the people."
Gary Barnett, an analyst with Ovum, said this is a great idea, not just for SilverHook, but for other racing outlets.
"It's definitely an interesting use of the cloud, for sure," he said. "And in the context of racing, this is really significant. In Formula 1 racing, the ability to get real-time data from the car to the engineers during a race has become crucial in winning races.... The big benefit of basing this on cloud infrastructure is the idea of what's next? Having designed the solution this way, SilverHook can easily add another boat or boats. They just scale the infrastructure to support more data."
Silverhook is working with IBM to build their private cloud on Bluemix, IBM's cloud platform.
Allan Krans, an analyst with Technology Business Research, called it "a good cloud use case, since it's a highly variable workload.
"During races, it needs high capacity," Krans said. "Afterwards, the traffic dies down very quickly. So this type of application is probably one that would not be feasible in a traditional IT environment."
Using a private cloud, as opposed to a public cloud, also helps SilverHook.
"One thing that's interesting ... is the use of a private environment, not for security because it's not sensitive information, but for performance and reliability," said Krans. "There's likely advertising and merchandise sales that would have a business impact if the feeds weren't working on race day and that justifies private versus the public cloud."
The big issue for Silverhook has been reliably transmitting all of the boat's data to the cloud while the boat is doing more than 140 miles an hour and being sprayed with salt water. That makes holding a connection and reliably transmitting a live stream of data -- using a basic SIM card and a cellular network -- a big challenge.
Hook noted that racers can get decent cell coverage a mile or two out into the ocean; otherwise, they switch to VHF radio.
"You need a real robustness of the physical performance of the entire system," said Hook, who noted that work on the boat's cloud system started in the spring. "At the speeds we're going and the impacts we're taking, you have to ensure a constant stream of data to the cloud. If a boat hits 6.9 Gs, there could be an interruption to the data stream. If it's being transmitted six times a second, then the analytics software IBM has can join the dots. If there's a gap in the data, it can propagate the missing data in between. That keeps a steady stream."
Brad Shimmin, an analyst with Current Analysis, said it's extremely important to be able to trust the data and the analytics the crew chief is seeing. That makes it important for IBM to reliably fill in any blanks that occur.
"Given some of the technologies IBM has been building, like Watson for predictive analytics, I can see them being able to really understand the data set," he added. "They can look at the data and be able to derive and predict [any missing] data. They can say, 'This is what we should be seeing.' Think about how invaluable it would be to have real-time streaming data off of these systems to apply predictive analytic algorithms to counteract potential problems with the network or the uplink."
That, according to Shimmin, will be the big proof point for this system.
"I don't think it tests the cloud itself," said Shimmin. "What it will point out is how difficult it is to have the scalability and the throughputs, no latency, no jitteryness, no lag for highly mobile sources of information. It will show us how difficult it is to pull this off the network that sits between the Internet of Things and the cloud itself."
That's important to companies that want to stream data from jet engines in flight, from ambulance companies and from shipping companies.
"It points out the difficulties of moving large amounts of analytical data from point of origin to point of insight," he added. "There's so much data to get to the cloud."
http://www.computerworld.com/article...the-cloud.html
http://www.silverhook.com/index.html
