The CPU can also execute instructions exceeding one per clock cycle. For example instructions such as MOV, CMP and even conditional jumps run at a typical throu

The CPU can also execute instructions exceeding one per clock cycle. For example instructions such as MOV, CMP and even conditional jumps run at a typical throughput of 3 instructions per clock cycle on the intel Core line of CPUs. (source: http://www.agner.org/optimize/instruction_tables.pdf http://www.agner.org/optimize/instruction_tables.pdf) This means that the simplest handcrafted loop that loads a byte, compares its value to the first byte of the searched string and then does a conditional jump should (if unrolled) take about 1 clock cycle per byte examined! This means that if our DDR3 memory can read 6GB of data per second and our CPU core is clocked at 3Gh, this completely naive algorithm will run at half of the theoretical maximum speed. Using XMM instructions (that work on 16 bytes at a time), should probably get us to the limit of the RAM speed. Regarding SAN-over-10-gig link. I don't know about you, but the computer I'm typing this on has an SSD that can read only a few hundred megabytes per second.