1. Designing a 300ms P99 Budget for a Public API Gateway
API ArchitectBackground
An architect needs to define latency limits for a new public API gateway to ensure the overall P99 response time stays under 300ms.
Problem
Manually calculating and visualizing how much latency can be allocated to network handshakes versus backend processing.
How to use
Set the total latency budget to 300ms, select the P99 SLO, and input the estimated stage allocations for DNS, TCP, TLS, TTFB, Processing, and Network.
Budget: 300
SLO: P99
Allocation:
DNS=5
TCP=10
TLS=20
TTFB=80
Processing=120
Network=40Outcome
A visual waterfall chart showing 275ms of the 300ms budget used, leaving 25ms of headroom, with no individual stage exceeding the 35% soft cap.