Site Reliability Engineering aims at improving the overall resilience of services across all applications and infrastructure. It embeds DevOps practices to bolster reliability without slowing the velocity of delivery. In order to achieve this goal SRE's monitor and measure metrics like Availability, Response Times & Error Budgets. These metrics become the eyes and ears that help define success of software products and eventually - business success.

This story is a DevOps Engineers account of how Redis turns out to be a game changer when it comes to performance and reliability of a high traffic serving microservices stack.

-- Prelude --

Once upon a time, there was a software engineering team who had undergone digital transformation. So from the rusty monolith software stack they now had shiny new microservices, CI/CD pipelines, Multi AZ databases, automated deployments, Kubernetes Orchestrators and Cloud-native serverless functions. Wow! sounds like a fairy tale. Not that easy!! Customer service team says, "user experience" isn't good, the application is slow, we have retry calls that slow down the backend. Now we the DevOps guys deep dive and see the DB disk I/O has spiked, there are locks waiting to be freed. SRE metrics start looking bad Error Budgets will be crossed, SLI's are screaming and SLA's might be breached. ((Panic Button!)) ...

-- End of Prelude --

In this session, let's discuss SRE metrics, the challenges faced by software teams who are new on the transformation journey to microservices and how Redis augments Site Reliability Engineering.