finextra, Pantor sets Fast protocol benchmarks, here. 2006 press release.

FIX Protocol, FPL Technical Roadmap for 2011, here.

   Andersson Pantor: Slides seem competitive technically and tight to reality. For example, the latency map to 2011 system architecture seems plausible:

100 us  50 us  10 us  5 us  1 us …

 100 us (one hundred microseconds)  Tuned OS and application logic, optimized GbE network

 50 us (fifty microseconds)  Highly optimized application logic, real-time OS configuration

 10 us (ten microseconds)  Kernel bypass comms drivers, low memory footprint

 5 us (five microseconds)  10G Ethernet, cache-aware application logic

 1 us (one microsecond)  Hardware assisted implementation?

and Andersson is presumably showing average test message sizes:

Test results  FIX tag value  Binary tag-value  Simple FAST  Full FAST

~200 bytes ~100 bytes ~35 bytes 15-25 bytes

4.6x compression w. FAST (not entirely sold on the compression idea architecturally, but you can compress and decompress in hardware reasonably fast I suspect)

< 100ns CPU overhead encode/decode per FAST formatted message (we were looking for ~60ns previously with Sandy Bridge)

   Henry Young TS-Associates: Great diagram pg. 50 in the FPL Technical Roadmap; Buy Side, Broker, Venue schematic. See ts-a.com, here.

About these ads