Great. I've thought about sharing ignore lists more than once. Might put together a greasemonkey script...
Port roulette at the CME "port roulette" refers to the assignment of ports to customers by the exchange, in an environment where not all ports have the same hop-distance from the source server(s) and/or the same network hardware. When you request a new port, it's a gamble (the "roulette" part) whether you get a port with good latency or not. http://stackoverflow.com/questions/...t-hft-trading-systems-today/22082528#22082528
"...Since a millisecond advantage in trading applications can be worth over $100 million a year to a brokerage firm (See article below – Business at Light Speed), this is a small price to pay. According to the NYSE, round trip latency to matching engines for colocation users in the Mahwah facility is 75 microseconds. The NYSE keeps this latency consistent throughout the facility by running the same length of wire to every rack in the building regardless of the physical location of the rack. This keeps trading firms from fighting over where their servers are located in relation to the matching engines. Another interesting piece of trivia . . . The NYSE does not allow carriers into the building. In order to connect to the NYSE you must first connect to the SFTI network in one of the buildings where SFTI is located, like 111 Eighth Avenue, 32 Avenue of the Americas, 165 Halsey, and a few others...." http://blog.compstak.com/high-frequency-trading-and-the-nyses-data-center-in-mahwah-nj/
Unless there is an independent audit of their switching fabric then consistent cable lengths mean squat. They could still incentivize by prioritizartion of traffic post cable.
I think all of this talk of latency is B.S. unless there is a terrific P/L ratio for the algo and a decent avg profit / avg loss ratio as well. A lot of algos can be profitable with the speed advantage....untill the commissions are substracted.