Cost efficiency, speed, and engineer wellbeing: How the growing fintech Delio simplifies its Kubernetes management
Helping a fast-growing AdTech platform Boostr to gain full control over cloud resources
A successful startup, Boostr grew quickly and saw a great deal of complexity creeping into its operations, with cloud expenses skyrocketing as a result. The company was looking for a solution that would reduce costs by resolving that complexity and allow the team to manage resources better. By implementing CAST AI, Boostr runs applications in an environment that is easy to manage and fully visible, offering many opportunities for eliminating cloud waste and cutting costs.
New York, NY
Cloud services used
The cloud is a must-have for every SaaS company
Boostr developed a platform that brings unified visibility to support sales, ad operations, and finance teams, helping them to scale omnichannel ad revenue profitability thanks to custom workflows, actionable insights, and accurate forecasting.
But the cloud comes at a price
As the company grew, the demand for the so-called sandbox environments created for every new feature increased dramatically, leading to a massive increase in cloud bills. The available resources weren’t being used to the fullest and servers were kept running during the low-demand time (outside of working hours and on weekends).
Boostr started thinking about reducing the cloud bill, considering the future scalability of both the team and the product itself. Migrating to Kubernetes seemed to be the right solution but the company lacked the required experience with the container orchestrator. This is where CAST AI came in.
Solution: Migration to managed Kubernetes with CAST AI
Since Boostr used virtual machines, its team had to separate all the elements of the development environment into separate containers to make them manageable via Kubernetes.
The CAST AI team assisted Boostr in separating the code to create a deployment system that was easy to manage and enabled a streamlined CI/CD process. The company started using the CAST AI platform to smoothly deploy the sandbox environments.
Result I: Gaining full control over cloud resources
Boostr product teams no longer deal with overloaded sandbox environments. If a developer wants to build a new feature, they don’t need to ask a DevOps engineer to create a sandbox environment and give them access to the server.
The developer can use Jenkins to create it automatically. There’s no need to wait for DevOps to respond or get approval for additional resources because CAST AI takes care of the node provisioning automatically. All the pre-production sandboxes are deployed with CAST AI, which allows the team to burst above the set number (40 sandboxes) whenever necessary.
When sandboxes are not needed anymore, they can be quickly destroyed to save capacity and costs. CAST AI also allows pausing and resuming clusters when nobody is working on sandboxes, saving additional costs without any added complexity.
Thanks to CAST AI, Boostr reduced its database restore time for new slots from 30-60 minutes to just under 30 seconds with the help of Kubernetes PVC snapshots across namespaces.
The setup didn’t disrupt the work of product teams – the implementation of CAST AI was a non-invasive process as it doesn’t add any complexity into the application lifecycle or translate into any extra effort.
Result II: Automated cost optimization
As Boostr prepares to enter the next stage of automated cost optimization with CAST AI, the new setup has already proven to be more efficient. Since clusters have a fixed number of nodes, the team knows how many sandboxes can be created – thanks to Kubernetes, developers can create more than 2x sandboxes than previously for the same price.
Get results like boostr – book a demo with CAST AI now
CAST AI features used
- Spot instance automation
- Real-time autoscaling
- Instant Rebalancing
- Full cost visibility