Using abstraction to eliminate the burden of config and infrastructure

Pulumi and Terraform are incredibly powerful, but does anyone else sometimes feel exhausted by the amount of time you still spend writing infrastructure? In situations where you don't need or want as much control over your environment, abstraction can help you move fast and not worry about config or infrastructure.

Here's a simple example to show how easy it is to write serverless applications with the Nitric framework's abstraction layer.
Coupled with a powerful interpretation layer this code utilizes an abstraction layer to self-describe the infrastructure, resources, and policies required to create a scheduled service that periodically deletes all files in a bucket.

import { schedule } from '@nitric/sdk';
import { bucket } from '@nitric/sdk';

// A schedule that cleans up files in a bucket every 3 days
const reportSchedule = schedule('cleaning-day').every(
  '3 days',
  async (ctx) => {
    const assets = bucket('assets').for('deleting');
    const files = await assets.files();
    await Promise.all(files.forEach(async (file) => 
        await file.delete()
    ));
  }
);

That's all you'd need to write. And automatically, all of the following resources are provisioned from the same code base with zero-config (almost … we need a few lines of config to tell the Nitric CLI where to deploy).

Let me illustrate what this means for whichever cloud you choose.

AWS

GCP

Azure

Curious to hear what you think of this use of abstraction. Check out more about how it works here.