FinDevOps and Site Reliability in the Serverless Age
summary
Argues that serverless requires rethinking SRE and DevOps around concrete tradeoffs: introduces FinDevOps and treats cost as an operational metric, not an afterthought. Maps SRE practices (SLOs, embracing failure, runbooks) to serverless realities like AWS Lambda and API Gateway limits, shows observability patterns replacing traditional monitoring, and gives a cautionary example where an internship project racked up a $1,000,000 AWS bill. Recommends measuring cost per transaction, baseline idle cost, and designing for service limits and rapid recovery.
AI-written summary · about
chapters (10)
- 0:00Introduction and overview
- 1:15Serverless as business enabler
- 3:44Rethinking cloud as operating system
- 5:50Defining reliability in serverless
- 7:58SRE changes in serverless world
- 10:04Observability over monitoring
- 12:13DevOps evolution with serverless
- 15:06Cost as first-class metric
- 20:38Aligning cost with customer value
- 23:09Future of FinDevOps
related talks
27mSite Reliability in the Serverless Age
41mDevOps in a Serverless context, opportunity [...] ?
05mUnleash your DevOps with Serverless
40mDevOps in the era of serverless computing
46mTwo Frames on Development and Operations
15mThe Ops in Serverless
29mHow is the role of DevOps changing in a Serverless environment?
05m#Serverless + #Devops = BFF