Sergey Nivens - Fotolia
FSB Technology (UK), which runs a betting platform service, has used GridGain’s in-memory database to accelerate its data processing.
The company operates in the businesses-to-business market, offering an online gaming platform. It began by focusing on fantasy football games, which integrated into other platforms. “We wanted to see what we could do with the interesting data coming out of sport,” says Sam Lawrence, who co-founded FSB Technology a decade ago.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
“Rather than just adding up scores at the end of a game, we tried using real-time data. For the last five years, we have been doing sport betting.”
FSB now provides a platform for online gambling and casino sites including 188Bet, BlackType online casino and bookmaker Toals in Northern Ireland.
In effect, it offers sports betting as a service. “Sports betting technology is expensive,” says Lawrence. “Data is becoming a bigger part of this cost.”
The speed of data has a direct impact on the company’s bottom line, he says. “With sports betting, we have the problem of data arriving into a platform that needs to be processed quickly. This data has to be processed fast as it will directly affect margins, so there is always a driver to make data processing faster.”
The company runs the open source PostgreSQL database as its main transactional data store. Lawrence says: “As the company grew organically, we chose open source software that we could get up and running quickly with a commercial support contract.”
But given the need for speedy data processing, he points out: “We needed fast complex query support and the ability to read and write data fast and to scale.”
The company has begun using GridGain, an in-memory database built on Apache Ignite as a data cache.
GridGain says its in-memory computing technology enables massive scale-out of data-intensive applications. It promises to dramatically improve transaction times compared with application architectures with disk-based databases.
Huge amounts of event data
At FSB Technology, huge amounts of event data must be updated constantly, and must be immediately available to a vast number of clients. For example, as a game progresses, new betting opportunities emerge. Odds must be calculated and presented to users in real time, and bets and event results must be processed instantly.
“We have progressed from application database architecture to a cache,” says Lawrence. Data is offloaded and kept in synch on a second data tier for fast in-memory database reads. “For the most part, we use GridGain for reading,” he adds. “On the risk side, when a bet is placed, we have to do fast calculations to ensure we are not exposed.”
FSB can pull complex data structures from its database rapidly and reliably. For example, if a better wants to see all current opportunities for betting on a particular sport, GridGain supports a single, fast query that pulls all the current odds for all the current bets for all the current events for that sport.
“GridGain is extremely flexible,” says Lawrence. “We have been able to create a great user experience for a variety of devices, providing the exact information users need when they need it.”
Although FSB started out using GridGain as a data cache, it is now being used as a layer between the database, says Lawrence. However, there has not been a complete shift to in-memory technology.
For instance, GridGain offers transactional support, but this is an area that FSB is not yet looking at. Lawrence is confident the company will make more use of in-memory technology going forward. “Having developers on the team who have knowledge of GridGain is a great way to get an understanding of new technology,” he says.
Private and public clouds
The company operates across both the private and public clouds, using Rackspace and Google’s public cloud, which means it can add capacity dynamically. This allows it to spin up extra GridGain nodes as and when extra data processing is needed.
Lawrence explains: “Generally, we spin up extra nodes at weekends or during a Champions League game.” This enables FSB to support the extra demands on its systems from peaks in betting, which often occur during major sporting events.
“GridGain’s ability to dynamically add and subtract nodes in the cloud has been critical to cost-effectively scaling our business and meeting our performance goals, even during extreme usage spikes, such as the Grand National,” says Lawrence. “And distributing our cluster across multiple datacentres has ensured availability.”
With GridGain, FSB also easily spun up another instance of the in-memory computing platform for a partner that wanted to run on a dedicated instance of the managed service.
In-memory processing is becoming more mainstream. At the end of last year, Stephen Hawking’s Centre for Theoretical Cosmology (Cosmos) announced it would be using HPE’s in-memory technology to further its research into the early universe and black holes. And in July last year, Japanese telco NTT Docomo rolled out SAP’s Hana in-memory database to analyse data from retail outlets and customer call centres in a bid to improve service levels.