Oddschecker: Global betting site solves scaling with a NoOps approach

About Oddschecker

Oddschecker offers an online odds comparison site, which aggregates betting prices for customers. Based in the United Kingdom, it operates in six countries worldwide.

Industries: Media & Entertainment
Location: United Kingdom

About Jetstack

U.K.-based consultancy Jetstack helps start-ups and large companies build cloud native infrastructure and specializes in the use of Kubernetes.

Betting aggregation platform Oddschecker uses Google Kubernetes Engine to power its B2B widget suite, scaling automatically to cope with spikes in demand and gaining efficiency through a NoOps approach.

Google Cloud Results

  • Autoscales to match spikes in demand around sporting events without additional infrastructure
  • Provides flexibility to trial new widgets without up-front investment
  • Empowers developers to maintain systems and work more efficiently

Serves widget data with sub 100ms response times

By aggregating prices from leading bookmakers online, Oddschecker helps customers choose the best price for their bet. Since 1999, Oddschecker has grown to be the United Kingdom’s leading odds comparison website. The company has also established a global presence with operations in 6 countries employing more than 80 staff.

“Being in a physical data center stifles our growth. The primary driver for moving to Google Cloud Platform was getting out from under that control, to be able to expand and try different things without requiring a large investment in physical hardware installations.”

Matt Robinson, Head of Technology, Oddschecker

Because Oddschecker features over 40 sporting categories and 100 different bookmakers, it handles very large quantities of data that have to be constantly updated to reflect changing odds. However, its legacy physical data center lacks flexibility and restricts how much data can be used and stored. So when Oddschecker decided in 2016 to develop a new B2B project, Oddschecker Connect, it seized the perfect opportunity to trial a shift away from physical data centers toward more flexible cloud infrastructure. With the support of cloud consultancy Jetstack, Oddschecker decided to develop this project on Google Cloud Platform.

“Being in a physical data center stifles our growth,” says Matt Robinson, Head of Technology at Oddschecker. “The primary driver for moving to Google Cloud Platform was getting out from under that control, to be able to expand and try different things without requiring a large investment in physical hardware installations.”

Building a Kubernetes-based platform

In order to deliver the best available price offers to their customers, odds aggregation websites need to provide real-time updates on the latest odds, even when demand spikes around sporting events. As Oddschecker relies on a physical data center, it was not able to easily scale to meet this demand and, as a result, its website experienced latency issues. The company wanted to use Oddschecker Connect, its new B2B project offering odds comparison widgets, as a test run for transitioning the entire platform and solving these issues.

To do that, Oddschecker decided to build its Oddschecker Connect infrastructure on Google Cloud Platform using Google Kubernetes Engine and Google Container Registry. Thanks to a data pipeline developed in partnership with Jetstack, Oddschecker was able to launch its B2B platform within five months from conception to launch, in time for a key event in the horse racing calendar: the Cheltenham Festival.

“Using Google Kubernetes Engine, we’ve moved towards a NoOps approach. The team has been able to add and improve this product with very little input from the platform infrastructure team. That’s why we’ve seen this platform evolve really quickly and to a very high standard.”

Matt Robinson, Head of Technology, Oddschecker

“It was a trial by fire,” says Matt Robinson. “We launched in partnership with a really high traffic platform, The Daily Mail, Britain’s widest read online newspaper.” The autoscaling function in Google Cloud Platform helped Oddschecker Connect easily meet demand, and the company did not need to invest in additional hardware months in advance, saving money.

In addition, using Google Kubernetes Engine and Google Container Registry, Oddschecker has developed new release processes and a new version control system. As a result, its system administration for Oddschecker Connect has been transformed, empowering its developers to take over system maintenance and experiment more.

“Using Google Kubernetes Engine, we’ve moved towards a NoOps approach,” says Matt Robinson. “The team has been able to add and improve this product with very little input from the platform infrastructure team. That’s the real benefit from my point of view, as it’s empowered the developers to be more autonomous. That’s why we’ve seen this platform evolve really quickly and to a very high standard.”

Jetstack: a perfect partner

In order to build its Oddschecker Connect infrastructure, Oddschecker partnered with Kubernetes experts Jetstack. Jetstack provided end-to-end support from conception to implementation, and further training. “We’re not a generic DevOps company that just puts teams on-site, we’re really there to upskill customers as quickly as we possibly can,” says Matt Barker, co-founder at Jetstack. “We worked on getting the engineers up to speed on Google Kubernetes Engine. The feedback from them was that Google Kubernetes Engine was intuitive to use.”

“Jetstack really helped us both from a high-level thinking point of view, and actually getting hands-on and helping to write and build the platform,” says Matt Robinson. “They were part of the whole end-to-end process, they were embedded within our team to help build the platform and then support us when we got to the go-live phase, to make sure the traffic and the platform were stable during the event we were targeting.”

“Oddschecker Connect was the testbed for getting everything into the cloud and testing it from a business perspective. It’s been a huge success, so we’re looking to transition our whole platform to the cloud over the next six months.”

Matt Robinson, Head of Technology, Oddschecker

Moving the platform to the cloud

Thanks to autoscaling, Oddschecker Connect served widget content for the Cheltenham Festival in 500 milliseconds to one second, dropping to 100 to 200 milliseconds when cached. Following the success of its Oddschecker Connect launch, Oddschecker is now planning to transition its whole system over to Google Cloud Platform to solve latency problems, move towards real-time processing, and develop new product streams.

“Migrating our current stack and engineering our platform to run on Google Cloud Platform effectively means we’re going to start using things like Google Cloud Dataflow and Google Cloud Pub/Sub, and really start leveraging some of the products that Google Cloud Platform offers,” says Matt Robinson. “We don’t have to build them all from scratch, we can just start plugging them in.”

Now Oddschecker hopes to use the powerful data processing tools on Google Cloud Platform to start gaining actionable insights from the data it stores, not only for its own use, but also for the bookmakers it deals with. “We’re looking at using Google BigQuery and Google Cloud Dataflow,” says Matt Robinson. “Because we work with a lot of data, being able to capture it, store it, and reuse it with analytics means we will add value back to our customers.”

“Oddschecker Connect was the testbed for getting everything into the cloud and testing it from a business perspective,” says Matt Robinson. “That’s been a huge success, so we’re looking to transition our whole platform to the cloud over the next six months.”

About Oddschecker

Oddschecker offers an online odds comparison site, which aggregates betting prices for customers. Based in the United Kingdom, it operates in six countries worldwide.

Industries: Media & Entertainment
Location: United Kingdom

About Jetstack

U.K.-based consultancy Jetstack helps start-ups and large companies build cloud native infrastructure and specializes in the use of Kubernetes.

Google Cloud Platform logo

12 Months FREE TRIAL

Try Kubernetes Engine, BigQuery, and other Cloud Platform products with $300 in free credit and 12 months.

TRY IT FREE
Google Cloud Platform logo

12 Months FREE TRIAL

Try Kubernetes Engine, BigQuery, and other Cloud Platform products with $300 in free credit and 12 months.

TRY IT FREE