Nasdaq to Begin Displaying Customer Latency

The enhancement to QView Latency Optics, the web-based tool released in February for tracking order flow, provides additional transparency into the Nasdaq data center infrastructure, so market participants can monitor order lifecycle and latency.

"Most people have figured out that latency variability is like the weather - sometimes it's hot, sometimes it's cold, and you just need to dress appropriately, provided you're given the information," says Donal Byrne, CEO of Corvil, which is supplying the technology.

Nasdaq QView users receive an overview of measurements in microseconds for the roundtrip of an order as it travels through the exchange network. Measurements are broken down into categories including: Order to Ack, which is the time it takes for the acknowledgement to be received back from the matching engine; Order to Book, which is the time it takes for the order to be displayed on the TotalView ITCH multicast feeds; and Cancel to Out, which is the time it takes for the cancel acknowledgement to be received. Firms may view overall session latency for the current trading day, timetables with interactive graphs for a granular examination of session flow, and percentile charts for each measurement category for a holistic view of the day's trading.

The tool may be used for diagnostics data as firms have the ability to compare their latency to that of their peers for a given connectivity method. Though individual peer latency is not visible, mean latency for a given peer group─1Gb network, 10Gb network, etc─is. In addition, firms may analyze trends utilizing latency categories by percentile and further specify by time interval.

Nasdaq customers had been asking for a clearer view of where their latency stood compared to other users, says Stacie Swanstrom, head of access services at the exchange. Byrne says Nasdaq is the first to offer such transparency.

"It's one thing to recognize that you're having a latency issue with your trading strategy, but what people really want to know is what the root cause is, and is there anything more sinister going on," Byrne says. "The only way to get that is to get the associated data with the anomalous event. There are an awful lot of folks today interested in understanding the orderly operations of their trading strategy instead of just speed."