One of my customer ask me to provide multiprotocol over CVO ( that´s great ) but including S3. The main reason for that it´s the cost of Persistent Disk on GCP .
My question it´s about the S3 bucket ( must be over CVO and not direct over Cloud Storage ) they need to provide the S3 but in the backend must point to the Cloud Storage bucket .
Reading the docs i think the only way it´s to tier the cold data to the Cloud Storage but we can not pass the S3 without using Persistent Disk ( aggregate level ) that´s correct ?
Main purpose of this it´s because an application only works with S3 API and doesn´t support Cloud Storage from Google and Persistent Disk are too expensive .
#S3 Queestion on CVO with GCP Cloud Storage
obsidian vector
steep carbon
So that basically sounds like they want to use CVO as a rather expensive S3-to-GCP-Cloud-Storage gateway? It appears the S3 API is capable of directly accessing GCP Cloud Storage to some extent: https://vamsiramakrishnan.medium.com/a-study-on-using-google-cloud-storage-with-the-s3-compatibility-api-324d31b8dfeb
