Financial trading institutions need a clear strategy for cloud implementation, writes Jeff Mezger, Vice President of Product Management at TNS Financial Markets.
Despite being a mature technology in many industries, the adoption of cloud computing is still relatively new in some sectors of financial trading. Traditional financial trading firms, from exchanges and data vendors, to brokers, hedge funds, and even proprietary trading firms, are racing to embrace cloud computing. For FinTech start-ups the cloud is the go-to, helping to drive innovation in financial markets.
The cost benefits of the cloud for trading and execution are clear. It significantly reduces upfront costs of setting up infrastructure. The lower costs of scalable ‘pay-as-you-go’ models allow smaller organisations to leverage robust security and infrastructural elements, established by powerful hosting providers, while themselves remaining agile. The approach of ‘processing power-as-a-service’ is easily scalable to meet the needs of different sized companies, eliminating costly investments to scale up. Another benefit of cloud computing for the international financial trading industry is that it can be accessed by the user from anywhere.
The right strategy
That said, firms need a clear strategy for cloud implementation. Public clouds can be cost effective to put data in and keep it there – however, taking data out is expensive. Many customers look at the cloud as something that makes business simpler and lowers costs, but firms still need to understand how to operate day-to-day, while simultaneously adopting different cloud offerings where appropriate.
Public cloud environments are great for big data analysis and to help financial firms store, move, and manage large amounts of data. It is also good for transferring large volumes of data between regions, like London and New York, and can support adherence to different regulatory requirements, for example, by enabling real-time monitoring of trading activity.
It must also be noted that the cloud is not one general thing. There are different factors to consider, especially geography. Different providers have physical locations from where their cloud solution operates, so understanding that there is a geographic element to any cloud offering is fundamental. Most trading strategies require a highly resilient and scalable infrastructure. Many factors can affect low latency trading, especially hardware location, the number of network hops, deterministic network latency, the hardware specification, and the resiliency of an infrastructure’s architecture, including both the scale and power of network connectivity.
The nature of cloud-delivered services enables ‘dial-up’ resources on demand, ensuring operational resilience, but are products available specifically aligned to the demands of electronic trading? Such solutions need to be built from the ground up and focus on the performance and latency required for financial trading.
Large scale public cloud providers may currently struggle to accommodate the particular needs of low-latency electronic trading and co-located infrastructure required by electronic traders. So, typically traders have used public cloud for less latency-sensitive purposes, like risk management, data storage, development work and modelling new strategies.
Traders also need reliable connectivity and market data feeds to drive their trading. They need to look to work with companies specialising in building shared networks and colocation footprints, offering trading firms access to market data and trading connectivity on demand. By using these shared infrastructure services, traders get the benefits of expertise and technology, at a fraction of the cost of doing it for themselves.
Steps for financial trading institutions to consider moving to the cloud:
- Prepare data requirements: Understand current data requirements, both real-time and static and be clear that these can be met in the cloud environment you are considering
- Carefully consider location: The physical location of cloud servers can be critical to success
- Map infrastructure: This includes understanding every system in the current deployed environment and whether it will translate to cloud
- Cost: Make sure you clearly understand the cost/benefit.
Ultra low latency trading is vital for algorithmic trading and trade execution speed is critical to ensure the best execution. The best low latency strategies depend on powerful servers housed close to exchange matching engines, be these in Europe, the USA or Asia.
Colocation services, especially access to servers processing close to the source of the input data, provides the lowest possible latency between input and response – and speed matters.
Traditional financial firms are racing toward enhanced cloud adoption. However, institutions need to be strategic when transitioning to cloud infrastructure, in order to maximise the benefits and continually review and revise their adoption decisions. They need to consider working with experts as when the technology evolves, they will understand the best options for continued success.