Camilo Hernández
Google Beta
“never pay for idle” concept is attractive
Easy onboard
Easy deploy
Reduce dev ops investment
Adds some developer independence
Architectures are inherently scalable
Vendor lock-in
The function itself is triggered by a typed, cloud-provider-specific event.
Testing and debugging become more challenging
new security concerns
built for long-running processes
Performance may be affected
You see an opportunity to reduce costs
You want to have some "hardware independence"
You don't want to worry about scale up and down manually
You want to monitor in a easy way your services
Cron tasks with small processing.
Event triggering (ex: msg)
inconsistent traffic
You can't live with cold starts
You want flexibility to move easily out of the cloud provider
You are not willing to automate
specific security concerns
specific gateway or networking integrations.
built for long-running processes
You need to develop something fast and it's your first time with lambdas
large applications with a fairly constant, predictable workload