I’ve spent a large part of my career in and around the low latency field. It is not uncommon to hear people say “oh but low latency is finished now”, in fact, they’ve been saying that since at least 2008, but is it really?
What exactly is Low Latency?
Latency in trading is the time it takes between deciding to execute a transaction and that transaction reaching the market. Thus, in a low latency environment, you are striving to minimize the time lost in non-business functions, whether by transmitting your transactions over long distances or through inefficient internal data handling. Anything that adds to the “tick to trade” cycle, the time it takes from a price update (tick) to be disseminated from a trading venue to the time the same venue receives your order needs to be understood and justified.
Why is Low Latency (so) interesting?
Low latency trading is ever-changing. While the challenge has not changed over the years (that is, to be faster than the pack), what was once innovative and differentiating is now passé. Where once you could make money by (simply) being fast, that luxury now only applies to the few who have the tenacity and budget to push the limits to the extreme. But if “raw” low latency is no longer so attractive, it means that the second wavefront, where the “pack” are concentrated, is still competitive and ripe for innovation.
In this age of AI, big data and cloud, new trading styles are emerging. Once upon a time mentioning “cloud” in a low-latency conversation would have got you laughed out of the building, but cloud trading services are becoming viable, and as they mature their latency profile will become a point of contention. Add to this the forthcoming 5G rollout, which aims to deliver high-speed, high-bandwidth mobile communications, it won’t immediately compete with fixed trading links, of course, but it opens new avenues of innovation and could perhaps herald a secondary (or is it tertiary) low latency boom.
I want a job in low latency technology too! What are the key skills?
Whichever of the many roles within a low-latency team you desire, you need to be a passionate technologist and a creative thinker. A degree of obsessiveness will help too.
For developers, a strong background in a low-level language, often C++, is expected, but fluency in multiple programming languages will be an advantage for those working with the traders and quants who may well be modelling in Python, Julia, R or other higher level languages. On the other side of the software layer, an intimate understanding of the hardware and operating system on which you are running is essential. GPUs and FPGAs are increasingly common in a heterogeneous trading system and understanding when and when not to use them is at least as important as knowing how. For ops and engineering, tuning the operating system to ensure the best performance (and least interruption) is imperative. Likewise, network engineers need to dive deep into their network, while also managing relationships with connectivity and hosting providers and evaluating new opportunities. In all areas, be an expert in what you have and keep an eye out for new networking technologies that may give a much-needed advantage. Consider the rebirth of microwave and more recently shortwave to see evidence of this.
In summary, if you are a passionate technologist with an eye for low-level detail, low-latency remains an exciting space to build a career!
Neil Horlock spent 20 years at Credit Suisse, most recently as a global architect for exchange connectivity within markets IT.
Have a confidential story, tip, or comment you’d like to share? Contact: email@example.com in the first instance. Whatsapp/Signal/Telegram also available.
Bear with us if you leave a comment at the bottom of this article: all our comments are moderated by human beings. Sometimes these humans might be asleep, or away from their desks, so it may take a while for your comment to appear. Eventually it will – unless it’s offensive or libelous (in which case it won’t.)