Numbers everyone should know.

I reproduce below Slide 24 from Jeff Dean's keynote (PDF) at LADIS 2009.

L1 cache reference0.5 ns
Branch mispredict5 ns
L2 cache reference7 ns
Mutex lock/unlock25 ns
Main memory reference100 ns
Compress 1K bytes with Zippy3,000 ns
Send 2K bytes over 1 Gbps network20,000 ns
Read 1 MB sequentially from memory250,000 ns
Round trip within same datacenter500,000 ns
Disk seek10,000,000 ns
Read 1 MB sequentially from disk20,000,000 ns
Send packet CA->Netherlands->CA150,000,000 ns