Best practice for reducing latency (NY Instance, APAC Users/S3)

My Setup:

  • Directus Instance: Hosted on a server in New York.
  • S3 Storage: Bucket located in the APAC region.
  • Users: Primarily located in the APAC region.

The Problem: Our content creators in APAC are experiencing high latency on all interactions, particularly with file uploads and downloads. My hypothesis is that this is caused by round-trip requests that must travel from APAC → New York → APAC and back again.

The Goal: I want to allow an APAC user’s browser to upload and download files directly to/from the APAC S3 bucket, using the New York Directus server only for authentication and creating the file metadata, not for proxying the data transfer itself.

What I Have Tried:

  • Relocating the S3 Bucket: Moving the bucket to a region closer to the NY server did not significantly reduce latency for the end-user.
  • Cloudflare Proxy: We ran into issues with Cloudflare’s 100MB upload limit and its incompatibility with the TUS protocol for resumable uploads.

My main question is: What is the recommended or standard way to configure Directus to enable direct-to-S3 file transfers to solve this specific latency issue? I am considering building an interface extension to manage S3 file interactions directly, but this seems complex and difficult to maintain (and I don’t know if it’s possible or not)