Skip to content

Huge Improvement. #1331

@dirtySnowman-I

Description

@dirtySnowman-I

I implanted the full mode and it is working properly but there is huge latency. To my understanding the main cause is the google dose not execute the codefull.gs in specific place but base load balance of google servers it could be all around world which cause huge latency between codefull.gs and VPS. To solve this we can implant something like a Cloudflare worker(Cloudflare worker may not work) between codefull.gs and VPS which will reduce latency from 1500-500ms to 60-30ms between VPS and google script. To my understanding your full tunnel mode has only latency issue (speed can be increased with more codefull.gs deployment). these are a list of what are causing it:

  1. Apps Script batch cycle = 2000ms : I have no idea.
  2. Apps Script → VPS trip = 1000 - 100 :
    . buy VPS near google data centers (EU-Central or US-East)
    . configure a domain for VPS and set IPV6
    . Use Bunny.net's Edge Storage to Cache Tunnel Metadata
    . Deploy Cloudflare Workers + Durable Objects (Cloudflare Durable Objects support persistent WebSocket connections but would require wrapping your tunnel protocol inside WebSocket)
    . Use Fastly's Compute@Edge (Fastly actually supports TCP passthrough via their Compute@Edge platform but it may be expensive)

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions