In an uncertain economy, traders are looking for the most cost-effective ways to enter new markets. In this blog, weconsider the strategy options for trading success.
To date, the obvious options for cloud providers have mostly focused on the public solutions offered by Amazon, Google and Microsoft. While TNS provides access to all of these, are they the best fit for low latency trading solutions in the financial services industry? CME Group, the world’s largest derivatives exchange, has moved its technology portfolio over to Google Cloud Platform (GCP). CME selected Google Cloud to increase developer productivity and software flexibility, while ensuring the highest levels of security.
Google and Microsoft’s public cloud offerings can be effectively used for storing large amounts of data or for financial markets back office activities that are not performance sensitive. However, what about processes that are performance critical? Working with a company that has a presence in the key global data centers is vital. Players need to be in colocation zones, because if they are not, the provider cannot guarantee low latency or handle high volumes of multicast market data traffic – all of which are key for trading success.
Some new trading participants may consider setting up in datacenters themselves, but with current long supply chain lead times, securing, deploying and installing the hardware can result in major delays – switches alone are currently taking up to a year to be delivered. A managed service provider can ensure the best of all options and avoid the issues associated with ‘going it alone’.
Optimal deployment models
Behind every financial transaction is a labyrinth of algorithms and networking infrastructure technology, which needs to be understood and optimized to ensure customers execute an effective electronic trading strategy. This infrastructure management, connectivity, access to a market data feed and location requirements are all necessary to successfully participate in the highly competitive trading marketplace. In addition to TNS’ current managed trading service offerings, the organization is adding an inventory of highly tuned TNS-owned trading servers. These have been specified to deliver premium cloud performance with the security, speed and reliability that performance sensitive banks, buy side firms and proprietary trading firms require.
TNS has spent a lot of time analyzing the best server components and set-up options to reach what we believe is the perfect trading server model and we are going to deploy an inventory of these servers around the globe. From Q2 onwards, if a customer needs a server or multiple servers rapidly, we will have inventory on hand and will be able to install servers as quickly as a couple of days.
Low latency is vital for algorithmic trading. Many factors affect latency, especially hardware location and the resiliency of an infrastructure’s architecture. This also includes both the scale and power of optimum network connectivity. Trade execution speed is critical to ensure the best market prices are capitalized on, and a key competitive advantage comes from having the best communication links to hardware in premium locations.
The best trading strategies also require highly resilient and scalable managed infrastructure services, with powerful servers housed close to exchange matching engines. Processing close to the source of the input data provides the lowest possible latency between input and response – and speed matters. TNS’ ultra low latency trading Layer 1 technology for direct exchange access within the data center was the first of its kind to be offered and deployed globally, and remains one of the most advanced solutions in the market. It eradicates the need for multiple switches by using a single-hop architecture to deliver direct exchange connectivity in as little as 5 to 85 nanoseconds – impressive when you consider that the human eye takes 400 nanoseconds to blink.
Firms need to futureproof their trading infrastructure by working with a provider that has experience in managing access to vast amounts of raw market data, that can support multicast requirements, and that is able to offer scalable solutions to accommodate the demands of ever-expanding bandwidth.
This server leasing model is just phase 1 of TNS’ new offering and has been led by demand from banks, software vendors and buy-side firms. When these customers identify new trading opportunities, they need to move quick to capitalize on them. This is a natural evolution for TNS, with the best technology, world class engineers and the global reach to guarantee trading success.
Jeff Mezger is Vice President of Product Management at TNS with responsibility for its managed services for the financial industry. He oversees product development and strategy for market data, online and data center services.