Co-Location on Industry Radar
With data centers having supplanted the trading floor as the place where most trades get executed, the issue of co-location is receiving scrutiny from market participants and regulators.
“For several years, co-location has been a major focus of brokers, market makers, and proprietary trading firms offering high-speed trading to their clients,” said David Zinberg, principal, financial services & insurance – capital markets at Infosys Ltd. “Exchanges who own the data centers have been profiting from offering colocation services. So latency is now measured in microseconds instead of milliseconds.”
In the U.S., two big names in setting up high speed networks to trading data centers are Equinix and Savvis, but there are many players. Equinix, for example, hosts the data centers of several different exchanges in their locations in Secaucus and Jersey City (both in NJ).
Regulators have also taken a keen interest in co-location. A proposed rule by the Commodity Futures Trading Commission, issued in 2010, would require that co-location and proximity hosting services be available to all qualified market participants willing to pay for the services.
Such services could not be offered on a discriminatory basis to only select market participants or to select categories of participants.
For example, if the availability of a service became limited, thereby leaving some market participants or third-party hosting providers without adequate access, the CFTC would not view access to those services as open and fair.
A provision relating to “latency transparency” would ensure that information concerning the longest, shortest and average latencies for all connectivity options are separately and readily available to the public on regulated trading markets’ web sites.
“Co-location is the most common method of overcoming physical distance for high-speed trading. I.e., if you are located in Europe and you want to trade with minimal latency on Nasdaq, you co-locate your server in a data center, which is either in the same facility or as close as possible to Nasdaq’s trade servers,” said Zinberg.
But co-location and the high-frequency traders (HFTs) who use it may become a victim of its own success. “These technologies can cause major disruptions in the markets that impact the average investor,” said Zinberg. “In the coming months and years HFTs and exchanges will no doubt come under much greater scrutiny and regulation.”
The International Organization of Securities Commissions (Iosco) has also taken up the question of co-location in its Principles for Direct Electronic Access to Markets (2010).
In that report, Iosco noted that the fairness of latency differences resulting from different technical connectivity options, and in particular from co-locating high-speed algorithmic trading systems adjacent to exchange servers, raises significant technical and market integrity issues.
Market participants emphasized the need for fair disclosure and equitable access for all participants, Iosco noted.
However, the fund declines to provide it.
Algorithm leverages machine learning to find inter-listed liquidity.
Buy-side wants new and better ways to find block liquidity.
New algorithm generates IOIs to find chunks of stocks.
New algo uses conditional orders and IEX routing to help traders.