Questback is an online digital survey and market research company that sought to re-architect its inefficient monolithic application architecture to make it easier and more cost-effective to innovate.
After researching its options, Questback’s IT and leadership teams initiated on a plan to reinvigorate its IT architecture to better serve customers, expand its business, and improve cost efficiency while taking a phased approach to application modernization (rather than re-writing everything at once). The plans called for a two-step approach, first for application containerization, and then migrating to a microservices architecture.
The company looked to leverage Docker and Kubernetes for phase one. Although Kubernetes simplifies container orchestration, the process of deploying, running, managing, and handling upgrades for Kubernetes is time-consuming and requires significant in-house expertise, which wasn’t readily available. In addition to simplifying the operations of Kubernetes, Questback was reticent about vendor lock-in and needed a platform built on unmodified Kubernetes so that they could switch vendors should priorities change.
The first step towards architecture modernization was to containerize their applications and decouple them from the underlying infrastructure with Kubernetes. To speed and simplify Kubernetes operations, Questback needed a management platform that ensured all deployed clusters were ready to run in production (incorporating security, high availability, backup, and disaster recovery) and at scale. Kublr provided all that out-of-the-box, easing Kubernetes adoption for the IT team.
In addition, with Kublr the IT team benefits from automated and pre-configured cluster performance monitoring and alerting, event analysis, smart recommendations, troubleshooting, and preventative actions to rectify performance issues. Centralized logging and monitoring, and the ability to manage multiple clusters from a single control plane, vastly simplified their daily work.
Two other aspects played into the platform selection. With plans to run applications on Amazon Web Services (AWS), Microsoft Azure, and in a private cloud, Questback needed the ability to easily move applications between infrastructures when required—hybrid and multi-cloud deployment capabilities were a pre-requisite. Additionally, they sought to benefit from Kubernetes’ latest features (e.g., auto-scaling metrics). The fact that Kublr is built on top of vanilla Kubernetes was a significant plus.
Kublr has enabled Questback to address key challenges while re-architecting its infrastructure. From an operations perspective, Kublr automates all management and operations functions so that operations teams can focus on other business-critical functions.
Using the same technology stack and deployments for application-level logging and monitoring, normally complex tasks now only require a few tweaks rather than months of reconfiguration. Because the shared stack is enterprise-grade, built-in tools are ready-to-use, and configuration within Kublr is simple, taking weeks, not months.
With Kublr, Questback had its first application up and running on the new modernized architecture within a week and has shifted to the speed and agility of a continuous deployment (CI/CD) model, versus deploying every six months. With its speed to innovation increased, Questback is able to more efficiently outmaneuver its competitors.
In addition, the company realized significant cost-savings because of auto-scaling and the use of computational resources on-demand. In paying only for capacity used, Questback reduced its monthly cloud capacity by 2-3 times, yielding savings of approximately 40% in operational costs per year. Furthermore, with Kublr’s efficiencies, the company has reduced the incremental cost for each additional customer, which previously increased in a linear manner.
One key to making it all work was bringing in the Kublr team to assist with the work and to “give us a new perspective into how we should approach this technology-wise,” said Questback CEO Frank Møllerop. “That’s what we did, and that was one of the key reasons we were able to finish on time and on budget.”
- Automation of manual tasks, powering innovation
- Simplified infrastructure with centralized management
- Reduced cloud capacity usage
- More competitive product cost structure
- Cost and resource savings
- No risk of cloud vendor lock-in
- Faster deployments through CI/CD
- Responsive to customer needs