Jan-Lukas Else

Thoughts of an IT expert

Scaleway Serverless Jobs

Published on in 💭 Thoughts
Updated on
Short link: https://b.jlel.se/s/746
AI generated summary: I explored Scaleway's new Serverless Jobs by creating a serverless Telegram bot that sends me AI-generated messages. The process involved coding in Go, using Docker, and scheduling the job. Serverless Jobs offer a longer maximum duration than AWS Lambda, running up to 6 hours. I'm considering it for backups but am weighing its benefits against my current use of GitHub Actions.

Scaleway continues to build a European AWS alternative. Today, I tried one of its new products: Serverless Jobs.

As I wanted to try this new feature and also play a bit with the OpenAI API, I built a simple serverless Telegram bot. The bot sends me a specific AI-generated message every morning.

To accomplish this, I created a simple Go project, added a Dockerfile and after coding, deployed the container image to a Scaleway Container Registry.

Then I was able to create a new job definition including selecting the memory and CPU size, as well as environment variables and a cron job schedule.

When the job is run, it pulls the latest container image, executes it with a specified command, and then terminates. To update the job’s code, I can simply build and push a new image.

The advantage over AWS Lambda is that the maximum duration is much longer. Instead of 15 minutes, it can run for up to 6 hours.

As I currently use GitHub Actions to back up two OneDrive accounts to Scaleway’s Object Storage, I am also considering using these new Serverless Jobs as an alternative. 🤔 But I’m not sure if it’s worth it, as I am staying well within the limits of the free GitHub plan.

Tags: , ,

Jan-Lukas Else
Interactions & Comments