NASDAQ-100 Enterprise: Scaling Cloud-Native Workloads
The Client
The client is a NASDAQ 100 company that specialises in human resources management software and services. The project was headed up by Jesse White, Contino VP Consulting (Financial Services) and Kubernetes expert.
The Challenge
Our client provides human resource management tools and batch processing to a large customer base. Historically, their IT estate has grown through acquisitions and highly siloed teams such as software development, systems administrations, operations, security and quality assurance. Our client identified the cost of running many disparate, decentrally managed systems. The client also lacks flexibility in their ability to provision and scale environments for rapid application prototyping, testing and deployment.
The Solution
Contino helped the client identify the core characteristics of a container-native, cloud platform that provided compute, networking and storage, while improving the elasticity, availability, and performance of their application workloads. Our chosen path was to build a Kubernetes Platform-as-a-Service on one of the major cloud providers, utilizing Cloud Native Services (CNS) such as Databases and Block Storage in order to complement the core competencies of the Kubernetes system. This allowed the client to create a universal set of concepts for teams across application development, security and compliance, quality assurance and testing and systems engineering.
By approaching Kubernetes as a framework of unified concepts, we were able to build a common language of application and infrastructure development, deployment and reliability. Key technical capabilities included: monitoring, cluster installation and package management, application development, CI/CD workflow, container image provenance, local development and host- and container-based security. We additionally helped the client re-architect their tightly-coupled, “macroservices” based system into a more modern architecture which utilizes key concepts such as a share-nothing datastores, common API standards, decoupled dependencies and key orchestration and scheduling concepts such as loadbalancing, circuit breaking, liveness, readiness and healthchecks. We’re setting the foundation for self-service infrastructure in order to take advantage of an SRE-type approach in the following quarters.
The Business Outcome
The environment provisioning time for infrastructure platforms has drastically decreased with the advent of fully automated, compliance- and security-vetted environments. By building in validation of controls into a templated, reproducible pipeline, the client has been able to onboard teams into new environments with a net decrease in time of over 75%. This enabled several key initiatives to go through development, testing and deployment to market ahead of quarterly goals.
By focusing on Infrastructure as Code, Kubernetes automation and cloud scalability, we were also able to reduce the personnel requirements for operating distributed systems. When we initially started the engagement, we assessed a 1:10 ratio of administrators to servers. At the end of this project we had increased the efficiency of that ratio to 1:100, and have goals of increasing that by another order of magnitude.
The third area of impact was in the net improvement to the security and consistency in environment configuration. This programmatic reproducibility has significantly reduced the workload on the audit, compliance and security teams and has enabled them to focus on creating automation and scanning tools of their own, allowing the client to push those concepts leftware into the Software Delivery Lifecycle. This approach lowers the cost of those improvements.
Across the organization Contino also worked with the client to incubate new ways of thinking in order to cope with the fastmoving, failure-friendly, fully automated approach. Through lunch and learns, frequent demos and in setting shared objectives and measuring progress through Key Performance Indicators and Key Results, we began to see demonstrable impact on employee job satisfaction and morale.
