The race to zero: Why latency matters in a MiFID II world


The trading game continues to evolve and technology is one of the key enablers behind the scenes. Computers have long replaced shouting traders in the halls of the exchange making it the fastest, not the loudest, that takes the prize.

For the world’s leading trading firms, especially those in the High Frequency Trading (HFT) sector, reducing network latency by nanoseconds is critical for increasing profits and competitive advantage. So it’s typical to see the major players invest heavily in ultra low latency infrastructure to connect their strategic data centres with trading colocation facilities at key exchanges.

MiFID II came into power in January of this year, and brought with it more stringent requirements for reporting and transparency, designed to make financial firms more accountable. The scale of the impact of MiFID II should not be underestimated, as the regulation will affect a firm’s trading infrastructure considerations on several levels.

One of the key requirements the legislation introduced is around timestamping – being able to accurately reference the time an event happens against a particular clock, not just within the event itself – something which can be a challenging feat when consolidating information from multiple sources on any given transaction.

The MiFID II definitions for timestamping and tracking of trade orders are set out in Article 25 of the directive’s regulatory technical standards (RTS 25) and introduce requirements for traceability and an accurate, sequential understanding of time. Timestamps have to be accurate to a particular clock rather than just accurate within the event itself. This means that the latency across a network matters for creating a consolidated view of network transactions – something that is deeply rooted in what the MiFID II requirements are trying to address.

The upshot is that latency has been pushed to the fore, as contending with these rules by introducing the kind of logging required, generally means a negative impact on the low latency components of the network.

There are several well recognised approaches to reducing latency at various points in the transaction chain, which can compensate for the increased overheads caused by logging requirements. Traders can move their own equipment closer to the source, such as an exchange colocation facility; they can reduce the number of pieces of equipment the data travels through; and they can use the shortest possible physical routes or lowest latency networks to get an edge.

For many traders however, such as those in the derivatives field, there is still a requirement to ‘normalise’ any data taken out of the exchange. Normalised feeds can be purchased at significant cost, which prices some players out of the market. Another option is to normalise data in software, but this approach is often too slow.

In a market where latency is critical, one option is to consume market data feeds normalised with an FPGA (Field Programmable Gate Array) solution. This hardware-based solution gives members and non-members of various exchanges access to normalised low latency market data feeds in a faster and more deterministic manner than software-based alternatives, enabling consistent, ultra-low latency performance through any market condition, available as dedicated or within a shared environment at a reduced cost.

Financial firms have to continually improve performance to retain their market share in a field where there is fierce competition and increasing regulatory attention. An FPGA-powered market data solution, combining ultra-low latency and flexibility without high upfront investment costs, is one way of overcoming this challenge and we have already seen a number of our customers express an interest in this type of service. Combined with a highly reliable network infrastructure, this approach allows all types of trading firms to benefit from low latency market data processing, in a simple and cost effective manner. This is for both members and non-members of exchanges and allows companies to shave valuable nanoseconds off their trading time.

Download the guide here.

Case Study

Colt PrizmNet case study: global asset manager

A single connection to Colt PrizmNet gives a global asset management firm access to the global capital markets ecosystem..

Related Product

PrizmNet Portal

PrizmNet Online, our self-service portal, is your easy-to-use interface to Colt PrizmNet.

Blog

Russia set to play a crucial role in meeting high bandwidth demands

Russia is primed to become a key connectivity route as the demand for global bandwidth continues to soar.

Call us

+44 (0) 20 7863 5510

Live chat with sales

For all your purchase enquiries.
Monday - Friday 9am-5pm

Chat with us

Contact our business sales team