Project snapshot
Our client, a global Fortune 500 manufacturer and retailer, has been building a worldwide ecosystem of AWS-based solutions to accelerate the time to market of their innovative flagship products. The staggering popularity of their electronic devices prompted our client to continue expanding their global presence. Intellias has been helping our client implement their business growth strategy by providing engineering and AWS cloud expertise to strengthen their technological infrastructure.
Operating in 180 countries and counting, our client has been a trailblazer in the consumer market with their revolutionary digital products. Through their next-gen customer engagement platform, our client has been delivering frictionless user experiences to 150 million users worldwide.
Business challenge
Our client’s global strategy called for fast scaling, increased stability, and high availability of their digital platform — all challenges that could be efficiently addressed using Amazon Web Services (AWS). But while the choice of cloud technology was obvious, the platform’s microservice architecture and complex infrastructure included many components developed across different teams that required multiple integrations and communication among systems. The goal was to orchestrate the behind-the-scenes processes and integrations fuelling the platform’s performance.
Our client’s rapid business expansion resulted in brisk growth of the customer base and a growing need to monitor user activity and track customer loyalty through QR codes. In addition to end consumers, a worldwide network of distributors, stores, and other retail outlets actively use the platform. Building robust, scalable, and optimized AWS cloud infrastructure would secure system stability, reliability of operations, and business continuity for our client, leading to sustainable revenue flows and more loyal customers.
Technology solution
Key project milestones
Our team of engineers kicked off work on the project by developing a basic product blueprint and choosing technologies together with our client’s solution architect. Our client’s platform has a microservices architecture and is an event-based system with Kafka at its core as the main service bus.
We started from two environments — production and non-production — and expanded to four major computing environments: development, QA, staging, and production. At first, our work was focused on enhancement and maintenance. Then we started building and adding new microservices to the platform.
At the current stage of the project, our team has taken full ownership of the development process and is involved in analyzing and estimating product features as well as engineering decision-making.
Amazon Web Services combo
The solution is built around a combination of AWS technologies including EC2, VPC, and Route53, allowing for a flexible and scalable architecture to ensure system stability and adaptability to increased loads.
- EC2 instances are used to support both an event layer (Kafka) and an application layer (consumers and producers). For applications, EC2 instances are managed by load balancers and auto scaling groups to adapt to loads.
- Different layers are isolated to their respective VPCs and are connected using Transit Gateway.
- Route53 is used for both monitoring and balancing requests in the application layer. With the event layer, Route53 is mostly static, enabling communications between Zookeeper and Kafka as well as Kafka and applications without relying on IPs.
Our team’s role
Because the platform is built in the AWS cloud, all integrations with the client’s other systems are smooth and easy to set up. For example, giving other systems permission to access the data center would be much harder without AWS. We’re also assisting our client with SSO compliance, performing SSO integrations with other applications and tools to beef up the platform’s security.
Our engineers use Kafka for real-time event processing to handle huge loads at maximum scale. We are responsible for managing the Kafka and Zookeeper cluster, handling the main load of messages (Kafka) and managing and keeping metadata for brokers (Zookeeper). Our task is to make sure all incoming messages are delivered without loss and consumed without delays. We use load balancers for EC2 instances to balance peak loads from one service across others and to perform scaling and resolve issues as the number of microservices is constantly growing.
Our experts keep tabs on the platform’s stability and performance by improving message delivery, enhancing processes, rebalancing loads, and adding retries and alerts — all to avoid long processing times and delays. We managed to cut message processing times from several weeks to a maximum of a couple of hours.
Currently, we’re migrating DevOps services to Kubernetes (EKS) and are planning to migrate the core business part of the solution as well. This will allow our client to cut costs for hosting and achieve more flexibility for scaling.
Business impact
Relying on AWS technologies, the Intellias team is helping our client build a secure, scalable, and performance-efficient platform with sophisticated architecture and infrastructure. We’re contributing to the system’s overall stability and availability across regions and platforms by improving system performance and speed, providing integrations, and optimizing infrastructure use and costs.
Given the massive scale and dynamic growth of our client’s business and the ever-changing demands of the retail industry, providing business continuity is critical to our client’s success. The platform that we’re building underpins continued business availability, performance predictability, and customer satisfaction, ultimately leading to substantial revenue gains and market share wins.
180
global markets covered
150mln
consumers worldwide
24/7
system availability
