This message was deleted.

Would be really great if you can share the DevOps practices and tools you are using to achieve this scale. I think that will help a lot of folks out here.

Hi Vivek. Thanks for taking an interest. In short we operate everything on AWS and use Terraform to define the infrastructure. We use managed services if possible and open source if managed doesn’t fit the bill and only opt to build if the needs can’t be met with either a managed service or open source. We use GitLab for all CI/CD including IaC pipelines. We’re not quite there but working towards fully codified deployments with as little human interaction as possible. Full automation is a little tricky due to contractual and compliance requirements, being in pharma. But Governance Automation is definitely shaping up to provide opportunities. The ratio is ever improving but now probably close to 70% of deployments are just a play button in a GitLab pipeline. My team runs the VPC, compute clusters (ECS and migrating to EKS, about 50:50 today), MSK for Kafka, RDS for relational DBs and so on. My team then provides monitoring, logging and alerting as services to the application/product teams (Prometheus+Grafana+PagerDuty and Coralogix for centralized logging, tracing being worked on) and initiatives ongoing to increase their DevOps maturity to be ever increasingly self-sufficient for deploying and operating their apps in all environments. It’s definitely a journey and the hard part is influencing people to change their relationship to their software (taking end to end ownership in place of throwing over the wall). The technology is comparatively simple to solve, the culture is hard part and the true driver of change.