Cloud vs Edge Latency Simulator

Network & Compute

Distance to cloud 800 km

Cloud processing time 40 ms

Edge processing time 8 ms

Request rate 4 req/s

☁ Cloud round trip

–

📍 Edge round trip

–

Edge speedup

–

How it works

Every request launches two identical packets at the same instant: one travels down the fiber to a distant cloud data center and back, the other is handled by the on-device edge node a few meters away. The cloud packet's delay is dominated by the round trip through the network — the further the data center, the longer light takes in the fiber, and every request the server is already busy with adds queuing delay on top. The edge packet skips the network almost entirely, so its delay is just the time the local processor needs to finish the job.

cloud_ms = 2·distance / c_fiber + processing + queue(rate)
edge_ms  = local_processing (negligible propagation)
speedup  = cloud_ms / edge_ms

Distance to cloud — kilometers of fiber between the device and the nearest data center; propagation delay grows with it even though light in fiber travels near 200,000 km/s.
Cloud processing time — how long the shared server takes to handle one request once it arrives.
Edge processing time — how long the local device takes to run the same job itself, with no network hop.
Request rate — more requests per second queue up waiting for the single distant server, adding delay that a local edge node barely feels.

Real-world relevance: this is exactly why autonomous vehicles brake on a local camera frame instead of waiting for a cloud response, and why CDNs and 5G edge servers push compute physically closer to users — shaving off milliseconds that a control loop, a game, or a safety system cannot afford to wait for.

Device Cloud request Edge request Cloud data center

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.