How to solve multi-cloud and deliver a great developer experience

When you’re handling over a quarter of a million transactions in 30 minutes, you need a cloud that operates at speed. When you’re looking to transform your business, you need a multi-cloud environment that doesn’t hold you back.

Ticketek is Australia’s largest live experience ticketing company. It issues more than 30 million tickets per year to a customer base of over 16 million users. And prior to the Coronavirus global pandemic putting ‘live’ events on hold, the business generated colossal amounts of data – where up to 300,000 tickets would be sold in one half hour period.

To hit those levels of transactional performance, this eCommerce platform uses the global cloud database service, MongoDB Atlas, running on AWS. However, in 2018 Ticketek launched a new secondary ticketing platform, Ticketek Marketplace. On the Marketplace,  customers can securely list purchased tickets for re-sale. Ticketek worked with Google Cloud Platform to build it. For the first time, it embraced a multi-cloud approach, running the platforms concurrently on two different cloud architectures.

Multi-cloud: advantages aplenty, but challenges too

Firms end up in a multi-cloud state for a number of different reasons. Some prize the flexibility or simply want to avoid vendor lock-in, while others are looking for the most cost-effective platforms to support certain workloads. The need for specific tools or integrations or differences in regional performance or pricing also comes into play – as does the pressing need for availability, uptime and business continuity in the middle of a global crisis.

For Ticketek, one key factor was that the Google App Engine service enabled its Marketplace to scale – and at speed too. This eliminates excess costs from over-provisioning while ensuring that the platform could still scale up to cover big events. It’s a great example of how adopting a multi-cloud strategy allows businesses to get the best from each cloud service, and to address their particular business needs.

Yet the multi-cloud approach does come with its own challenges. Multi-cloud is complex, and while you can scale and shift applications from one service to another, data is a whole different story. This is crucial for a firm like Ticketek where systems must handle thousands of simultaneous transactions, and where data is the key to growth and innovation; the company plans to focus on personalisation and targeted content to engage with customers and drive future sales.

The challenge is that it’s all too easy for data to become siloed between different transactional, analytical and search systems which is complicated further by data’s tendency to get caught with a particular cloud provider.

Pulling it all together from these different locations, be it from a public cloud provider, your own Hadoop data lake or one of a thousand other data sources, is no trivial task. There are complex dependencies to contend with, along with different tools and APIs. This often slows developers down, forcing them to focus on the infrastructure rather than the application. 

Unleash developer productivity with a Multi-cloud Database as a Service

Many businesses have found an effective way to traverse these challenges while still enabling their developers. It can be achieved by working with a Database as a Service (DBaaS) that has three key features. Firstly, it abstracts away underlying software, infrastructure, management and updates. Secondly, the DBaaS should be based on a developer-centric data model which can handle any data type and evolves at the speed developers think. The document model is the leader in this area. And the third element for multi-cloud success? The obvious one: The database as a service should be platform agnostic, so you can deploy anywhere, any time.

This is the choice Ticketek made with MongoDB Atlas. The firm now runs multiple MongoDB clusters in various regions across both the AWS and GCP clouds. These robust clusters can scale-up on demand. As MongoDB Atlas integrates with the full range of cloud technologies, Ticketek can develop applications and dashboards that use incoming real-time data for rapid insights into sales and trends. 

“I can’t stress enough how important it is for us to have access to reliable, point in time data,” says Ticketek’s enterprise architect, Tane Oakes. “Integrating MongoDB with Google Services provides us with the analytics necessary to personalise communication and deliver a better digital-first customer experience.” With the company expanding into Singapore, Malaysia, the Philippines and the UK, there’s scope for data to drive even more success.

Get started with MongoDB Atlas for free: https://www.mongodb.com/cloud/atlas