Skip to content

Instantly share code, notes, and snippets.

@notwedtm
Created November 18, 2025 12:34
Show Gist options
  • Select an option

  • Save notwedtm/034868c424d0f2f246bc65011b72934b to your computer and use it in GitHub Desktop.

Select an option

Save notwedtm/034868c424d0f2f246bc65011b72934b to your computer and use it in GitHub Desktop.
Latency Comparison Numbers (~2025)
Latency Comparison Numbers (~2025)
----------------------------------
L1 cache reference ~1 ns
Branch mispredict ~3 ns
L2 cache reference 4 ns ~4x L1
Mutex lock/unlock (uncontended) 17–25 ns
Main memory reference 70–100 ns ~100x L1
Compress 1 KB with Snappy (Zippy) 2,000 ns 2 us
Send 2 KB over 10 Gbps network 1,600 ns 1.6 us
Read 4 KB randomly from NVMe SSD 20,000 ns 20 us
Read 1 MB sequentially from memory 10,000 ns 10 us
Round trip within same datacenter 100,000–500,000 ns 0.1–0.5 ms
AWS AZ-to-AZ RTT (same region) 400,000–700,000 ns 0.4–0.7 ms
AWS Region-to-Region RTT (US-East ↔ US-West)
30,000,000–50,000,000 ns 30–50 ms
AWS Region-to-Region RTT (US ↔ EU)
70,000,000–110,000,000 ns 70–110 ms
Read 1 MB sequentially from NVMe SSD 100,000 ns 0.1 ms
Disk seek (HDD) 10,000,000 ns 10 ms
Read 1 MB sequentially from HDD 5,000,000 ns 5 ms
Read 1 MB over 1 Gbps network 10,000,000 ns 10 ms
Transcontinental RTT (CA ↔ NL, public internet)
150,000,000 ns 150 ms
Notes
-----
1 ns = 10^-9 seconds
1 us = 10^-6 seconds = 1,000 ns
1 ms = 10^-3 seconds = 1,000 us = 1,000,000 ns
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment