Estimate round-trip latency for distributed systems — propagation delay, serialization, processing, and multi-hop chains
Model a request path through multiple services (e.g., CDN → LB → App → Cache → DB)
| SLA Target | Downtime / Year | Max Latency Budget | Typical Use Case |
|---|---|---|---|
| 99.9% (3 nines) | 8h 45m | 500ms | Internal tools, dashboards |
| 99.95% | 4h 22m | 300ms | Business applications |
| 99.99% (4 nines) | 52m 36s | 200ms | E-commerce, SaaS APIs |
| 99.999% (5 nines) | 5m 15s | 100ms | Financial, real-time systems |
| Operation | Typical Latency |
|---|---|
| L1 cache reference | 0.5 ns |
| L2 cache reference | 7 ns |
| Main memory reference | 100 ns |
| SSD random read | 150 μs |
| HDD random read | 10 ms |
| Same datacenter RTT | 0.5 ms |
| Same region (cloud) | 1–2 ms |
| Cross-region (US) | 40–70 ms |
| US → Europe | 80–120 ms |
| US → Asia | 150–250 ms |
| DNS lookup | 20–120 ms |
| TLS handshake | 30–100 ms |
| TCP handshake | 1 RTT (varies) |