Achieving 99.999% Availability Using Amazon DynamoDB at Genesys

Genesys Cloud empowers more than 8,000 organizations in over 100 countries to provide a frictionless, personalized, and empathetic customer experience (CX) and employee experience while improving business agility and economic outcomes.

Traditionally, CX solutions were developed and installed on premises, requiring significant upfront investment in hardware and software. These systems were often inflexible and difficult to scale and lacked the agility that is needed to adapt to rapidly changing customer expectations.

Recognizing these limitations, Genesys set out to transform the CX industry by helping organizations evolve their operations to the cloud. Using AWS, Genesys can provide organizations with on-demand scalability, access to global AWS infrastructure, and the ability to innovate rapidly without the constraints of physical hardware.

To guide the development of its AI-powered platform, Genesys defined key principles to shape its approach. The team chose to build loosely coupled, tightly integrated microservices instead of a monolithic product suite to increase flexibility and accelerate updates. Genesys also aimed to minimize manual steps and operational overhead by automating every possible process so that it could reduce human error and increase efficiency.

Over the years, Genesys built hundreds of microservices that each perform a specific function. Each service contains three components: The first is a load balancer that efficiently distributes incoming traffic for optimal availability. The second is an automatic scaling group that is powered by either Amazon Elastic Compute Cloud (Amazon EC2), which offers secure and resizable compute capacity for virtually any workload, or AWS Lambda, a serverless compute service.

The third is a data tier, which is responsible for storing and retrieving microservice-specific data. This data tier is where Genesys uses Amazon DynamoDB, a serverless, NoSQL database service that empowers companies to develop modern applications at virtually any scale.

Deploying and managing individual database clusters for hundreds of services across multiple regions would have resulted in significant overhead costs and complex scaling challenges. Additionally, Genesys wanted to help its development teams maximize agility while minimizing the operational burden of infrastructure management. After exploring potential solutions on AWS, Genesys decided on Amazon DynamoDB.

Amazon DynamoDB can grow from zero to virtually any scale, with no cold starts, version upgrades, maintenance windows, patching, or downtime for maintenance. Therefore, the Genesys architecture team chose Amazon DynamoDB as the standard for any new microservice that requires single-digit millisecond latency at the data-tier level.

“Amazon DynamoDB is our primary data layer by default, and teams have to justify the use of something else,” says Rob Gevers, chief architect at Genesys. “There are use cases that fit better with a relational model, but we have significantly fewer of those than use cases that can be well served by a NoSQL approach. So far, we haven’t found any other database technology that lives up to the operational stability, scalability, and simplicity of Amazon DynamoDB.”