Cloudar Improves Speed and Availability for TinkerList on AWS

TinkerList provides an online collaborative script and rundown tool called Cuez for broadcasters, live event organizers, and esports companies. Users require high availability and low latency, but the company found that maintaining its existing infrastructure to such a high standard was proving difficult. TinkerList turned to AWS and AWS Premier Partner Cloudar to migrate to the cloud for improved performance, reliability, and faster development time. Now it no longer needs to provision systems for peak capacity but can scale automatically as demand changes. Compute costs are now tied to usage, so expenditure is more predictable and linked to revenue.

Belgium-based TinkerList provides a cloud-based content management application called Cuez that is used by TV and radio producers, online channels, live event organizers, esports companies, or anyone needing to collaborate on live media production. Cuez allows users to collaborate on editing in a script and rundown, automate and customize the media production process, and control live show timings.

In addition to scripts, the application can include video clips, images, and instructions for the production staff. Cuez can be used during live broadcasts to make last-minute changes and keep track of timings. The application integrates with other studio production hardware and software such as the teleprompting system Autoscript or with its own web-based teleprompter application, syncing all updates with the presenters’ tablets.

TinkerList’s existing platform had security, performance, and availability issues. This was a growing concern because the company’s users rely on it during live events, in which even a second of latency or downtime is unacceptable. In addition, the team was spending increasing amounts of time maintaining its systems.

TinkerList chose to re-engineer its platform on Amazon Web Services (AWS). The company lacked AWS expertise and needed a partner to help with development and ongoing operational management. TinkerList had previously worked with AWS Premier Partner Cloudar on a project at a Belgian broadcaster. After talking with other potential partners, TinkerList chose Cloudar for its responsiveness to queries and its pricing. The two companies formed a close working relationship, especially in the first few months of the project.

As TinkerList grew and added more customers, its service levels came under increasing pressure. “We had some hiccups managing it all ourselves,” says Erik Hauters, founder of TinkerList. “We wanted to sleep at night without worrying about infrastructure and to focus on our product and our customers. The old system had a server usage meter in the office and we’d watch it go from green to orange to red—it was a nightmare.”

Cloudar migrated TinkerList from a traditional infrastructure as a service (IaaS) platform to a cloud-native system. The redesign has cut costs and accelerated TinkerList’s ability to release updates to its application and increased its confidence that the changes will not cause issues for customers.

Cloudar systems now scale automatically using Amazon Elastic Container Service (Amazon ECS), a fully managed container orchestration service that simplifies deployment, management, and scaling of containerized applications, and AWS Lambda, which lets you run code without thinking about servers or clusters.

Using Cloudar’s managed services and AWS infrastructure means that TinkerList’s developers are free to focus on building its product and reacting more quickly to customer requests.

Credit: Tinkerlist

Expenditure on IT is now directly linked to customer use of the system and therefore to revenue. “Money comes in and money goes out, but it’s predictable and matched to our expenditure,” says Hauters. “We’re not spending money on computing that’s not being used.”

TinkerList’s previous system was expensive because it had to be specified for maximum load and peak usage. In addition, the system couldn’t be quickly downsized for quieter periods. Onboarding new customers also sometimes affected the service for existing customers.

System optimization work is ongoing, but Cloudar recommended that the company use Amazon EC2 Reserved Instances, which provided savings of €8,000 a year.

Using AWS, product development is also faster and testing is simpler. The company now has a dedicated environment to test new features and code releases. Previously, there was some dependence on customers testing releases on their own systems—given the broad reliance on third-party integrations—but new features can now be more fully tested and released without concerns about downtime.

TinkerList has so far not migrated any existing customer data to its new platform. New customers are joining the new cloud-native platform and the company will migrate customers from its legacy platform in the next 12 months.

The migration to AWS is also supporting TinkerList’s expansion in Australia and the US. The company is using AWS Regions and Availability Zones, which means that customers in those countries can experience low latency and TinkerList can provide the same level of service for customers anywhere in the world.

The project was made more affordable by the AWS Activate program, which provides credits and other support for startups. TinkerList and Cloudar are investigating how they can use more serverless services to further reduce costs. The two companies are also exploring the role that artificial intelligence could play in helping create scripts and running orders for customers.