FULL TUNNEL MODe I dont Care About the 20,000 hits that google permits ... can i make like 50 appscripts just to have Enough speed ?

[mitilimitan007]

Hello
I have tried one appscripts and i dont Feel like connected (its like Diging a grave with a spoon)
Now i was wondering can we deploy like 50 appscripts just to have enough speed ?
has anyone done that ?
like with 2 gmails make 100 Deploys and do we need to congfigure it in another way to make it use all of them ? (BTW im talking About the Full tunnel) My VPS is REady and Loaded
Thanks For Backing us UP Everyone Out there
GOD BLESSS YOU ! <3 <3

[dirtySnowman-I]

Did your full tunnel mode get connected? And regarding your issue, I think having more deploys on 1 account does NOT make any difference; Google handles them one by one (I think), but I can give you 100 Gmail accounts that will do the trick.

[therealaleph]

Reviewed via Anthropic Claude.

@mitilimitan007 — yes, you can deploy many Apps Script projects. Three constraints to know:

1. Cost vs benefit on Full mode: each script_id you add doesn't help speed proportionally in Full mode because Full mode is stateful per session — TCP sessions need affinity to one tunnel-node. Round-robin across many scripts breaks affinity. See #1163 closed yesterday with that exact scenario.

For apps_script mode (no VPS): adding more script_ids genuinely helps — each request is stateless, so 5 scripts ≈ 5× throughput on average.

For Full mode: 1-2 script_ids is the sweet spot. 50 buys you nothing.

2. Google account constraint: each script_id must come from a different Google account to get separate quota buckets. 50 deployments on 1 Google account share 1 quota — so you'd need 50 Google accounts. That's the actual hard wall (Google accounts are rate-limited by phone number, then by behavioral signals).

3. Speed bottleneck is not script count: at 1 script_id the bottleneck is Apps Script edge RTT (~200-400ms per call) + Iran ISP latency. 50 scripts doesn't reduce per-call latency. What reduces it: coalesce_max_ms higher (pack more ops per call), exit-node for CF-protected sites (cuts 1 layer), faster VPS region.

Recommendation: start with 2-3 script_ids from different Google accounts, set coalesce_max_ms: 3000, set exit-node for CF/x.com/chatgpt sites. Should feel meaningfully faster than 1 script.

می‌بندم — there's no "speed scales with script count" mechanism. Focus elsewhere.