The Blog of James: Do you use FIX or FAST? (Poll Results)
The January Poll at Low-Latency.com focused on FIX and FAST usage across the industry. The results are what many in the industry view as “normal”. See the graphic below.
Nearly three-quarters (74 percent) use FIX - that’s not surprising as it’s “the” de-facto standard for electronic trade communication (and it was the subject of the poll). But it does make one wonder what the other 26 percent are doing!
Nearly half (46 percent) use FAST - that all-so-useful “compaction” scheme primarily used for saving bandwidth on market data. It’s only a matter of time (I hope) that this scheme gets used for other data - keep up the good work!
And, it isn’t unexpected that there are a few stalwarts of days gone by (14 percent) that use neither FIX nor FAST. No doubt, they have a legacy system with sunk costs that does what they need it to do - for now. Or, perhaps they use somebody else’s tool or service that handles the “heavy lifting” of dealing with FIX and/or FAST. Fair enough - I’m sure their bottom line is sound.
But my primary point of interest lies with the 12 percent - a minority but significant enough - that use FAST without FIX. Methinks this group sees the advantage of compression - perhaps just for market data - but do not yet see the value of FIX for their needs. Or, stated differently, they use FAST outside the scope of FIX itself.
Or maybe FIX just doesn’t fit into the low-latency realm like FAST does? Perhaps the FAST proponents might offer an opinion?
Of course, this presents another problem - just as no two FIX-engines are alike and many FIX conversations are customized between the counterparties - we find that no two FAST implementations are the same. Perhaps not all of features of the FAST compaction scheme are implemented - thus one provider’s decoder does not work with another’s encoder. Or, some additional out-of-band communications or data have been added to the compacted scheme with would confuse a decoder from a different provider.
But, our observed increase in electronic trading via FIX can only be good - and the FASTer the better.
