If I had to pick the biggest trading infrastructure trend of 2024, it would have to be flexibility: how infrastructure can, and should, be adjusted to meet the demands of the markets. Not only has it dominated conversations with trading CTOs concerned about the challenges of having only one hosting provider, it is a talking point even the US’s self-regulatory organization for brokerages and exchange markets, the Financial Industry Regulatory Authority (FINRA), has been vocal about.
“Firms may consider the risks associated with vendor lock-in and the potential that cloud service providers might be unable to reliably provide services… firms may wish to consider whether multi-cloud or hybrid cloud options are compatible with their business needs.”
Increasingly, brokerages and regulatory organizations are reconsidering their use of certain infrastructures like hyperscale cloud and looking at taking more of a hybrid infrastructure approach to reduce their reliance on a single provider.
For those that are already in hyperscale cloud or considering it, FINRA has recommended the development of a cloud exit strategy in case they get to a point where they want to look at other options - “Alternatively, they may wish to consider adoption of an exit strategy to mitigate against an unfavorable lock-in scenario.”
While I don’t believe that hyperscale is bad and everything else is good, there is a clear shift in the trading world towards more reliable, flexible infrastructure that I am positive will continue into next year. But what other trends can we expect to see in 2025? In this blog, I cover my top four predictions.
The last couple of years have seen their fair share of proprietary trading firm controversy. Prior to last years’ amendment to SEC’s Rule 15b9, many proprietary firms were able to avoid registering with FINRA, meaning they were not subject to their oversight. Rule 15b9 brought to the forefront just how many proprietary trading firms (64 in total) were operating unlicensed.
While the rule change has meant a lot of extra work for previously unlicensed proprietary firms to become FINRA members if they want to keep trading, it has opened a new opportunity for brokerages to expand their offering with their own proprietary trading services. So far, OANDA, Axi, IC Markets and Hantec Markets are chasing that gap in the market, and I predict more are going to follow in 2025 for three key reasons:
Proprietary services offer customers a way of trading without risking their own capital. These services can be particularly attractive to novice traders, as they can gain access to an account worth thousands of dollars whilst only putting up a few hundred of their own.
By setting up proprietary trading desks, brokerages move beyond their standard commission earnings by participating directly in the market. They can take advantage of market fluctuations, profiting from asset prices as well as commissions from clients.
There is very little (if any) extra operational costs for brokerages to expand their offerings to include proprietary trading, as their infrastructure already possesses high-performance trading systems, data analysis tools, and risk management capabilities that can adapt to the new service.
On the flip side, proprietary trading firm FTMO is following the same logic, just from the other direction, by entering the brokerage space. Described as a “long term strategy”, offering brokerage services allows FTMO to generate a more stable revenue stream, unlike the often volatile profits of proprietary trading. These services widen their client base to investors of brokerages as well as to proprietary firms. I predict that other proprietary trading firms will follow suit if FTMO show signs of success with this new strategy.
The SEC’s introduction of stricter requirements for US proprietary firms has highlighted the need for greater regulatory control in the industry worldwide. MetaQuotes have been ceasing their operations with proprietary firms as a result, terminating contracts and licenses over security concerns. Italy’s security regulator Consob went as far as to describe proprietary trading as simulating “an online trading activity in a type of finance video game”, suggesting a sense of frivolity in what should be treated as a serious and regulated financial practice.
Since the development of these regulations in the US, it is not clear what proprietary regulations are going to be implemented in the rest of the world, but the industry is certain that changes are on the horizon: “There’s no doubt that when regulators start seeing issues, very often what comes out of it is a new regulatory framework” explains Remonda Kirketerp-Moller, Muinmos CEO. “We will see more regulation coming for proprietary trading firms, for sure.”
One regulation we do know will be introduced in January 2025 is the Digital Operations Resilience Act (DORA). This EU regulation focuses on digital compliance and cybersecurity, ensuring proprietary trading firms have a risk management framework. They will be required to monitor their ICT systems and tools, and to share their information on any cyber threats.
Compliance with DORA is expected to lead to “reduced financial losses from operational incidents, faster and more trouble-free implementation of new systems, maintenance of good customer service levels, increased brand value, lower risk management costs, as well as lower regulatory risk”.
I predict that by this time next year, advances in regulation mean that proprietary trading is seen more as a secure practice and less of a “finance video game” in the eyes of the regulators around the world.
The difference between a good trade and a bad trade comes down to microseconds. A low latency trading infrastructure is crucial for brokerages to ensure that the value of the trade is not lost during the transaction. Equally as important is how quickly trading firms can respond to market changes - adjusting their infrastructure to best cater to the demands of capital markets.
This need for speed and flexibility has typically been achieved by using virtual machines (VMs). VMs split up a server virtually, allowing different operating systems (OS) and applications to be run on a single piece of hardware. Using multiple trading systems seamlessly across one server has made VMs the ideal choice in keeping an infrastructure set up as flexible as possible.
However, because VMs operate at a hardware level, they intrude on the overall server resource such as memory and storage. This is where Linux containers come in.
Instead of virtualizing the server at a hardware level, containers virtualize the server at the OS level. They package applications onto the system’s kernel – the program that allows the communication between the hardware and software – and because of this, containers are extremely efficient. They cause no degradation to server resource and can be deployed much faster than VMs.
“This containers revolution is changing the basic act of software consumption. It’s redefining this much more lightweight, portable unit, or atom, that is much easier to manage… It’s a gateway to dynamic management and dynamic systems” says Craig McLuckie, Stacklok CEO and co-creator of Kubernetes.
Kubernetes is a platform that automates the deployment and management of containerized applications. According to Traders Magazine, “Kubernetes is now the de facto standard for container management and can be deployed on-premises or in the cloud using a kubernetes-management platform.”
Linux containers are a fast and resource efficient method to install different applications onto the same server, and Kubernetes can automate their deployments. For the volatile markets that traders operate in, I predict we will see an increased uptake of these practices so that their operations remain as adaptable as possible.
For brokerages in the start-up phase without their own in-house IT expertise, a one-stop-shop technology partner is a great way to go. They provide the necessary trading technology to enable a brokerage to serve its clients, whether that be customer relationships management, payment processing, liquidity aggregation, platforms, and of course hosting, either unmanaged or managed.
But for brokerages that are starting to mature and develop their own in-house technology teams, they can afford to move away from a single provider now that they have the expertise to manage multiple environments and relationships. By expanding beyond a single technology provider, they benefit from:
An over-reliance on one technology provider can cause brokerages to be locked-in to their services, unable to migrate away even when they see their costs eating more into their budget than anticipated.
In-house infrastructure expertise not only gives a business peace of mind knowing that they have the resources to tackle a problem themselves, but stepping back from a technology provider grants them more control over how they choose to keep their trading infrastructure secure. No longer at the mercy of third-party security systems, internal tech teams can create bespoke risk management that suits them best.
Being self-reliant doesn’t always mean doing everything yourself. Taking a step back from a single technology partner means a business has the freedom either to manage the technology stack themselves, or to source different technologies from a range of providers. The ability to pick and choose gives the highest chance of finding the best options.
Splitting an infrastructure environment out over multiple providers also creates much better redundancy, a need emphasized by a series of retail broker outages earlier this year.
As Kris Mullins, CMO of investment firm Capital Max commented on the outages, “these are more than just technical glitches; they are clear signs that much of the infrastructure supporting our financial markets is archaic and inadequate to support modern trading practices… a comprehensive overhaul is long overdue.”
Mark Hirsch, Founding Partner at Templer & Hirsch, adds that “brokerage companies should prioritize investing in their technology stacks… this improves service delivery and customer happiness, keeps things safe, and gives them an edge over their competitors.”
Brokerages diversifying their technology stack will offer them more protection against future outages, as they will no longer be vulnerable to a single point of failure.
But it can take a business years to acquire the confidence to develop their technology stack independently. As brokerages grow, their understanding of requirements grows with them. Whilst a single provider is a great starting point for those who are not there yet, it is becoming more prevalent - and crucial for security - for those who can to take a more diverse approach to their infrastructure. For these reasons, I believe we’ll see more diversification of technology providers at brokerages in 2025.
If 2024 was all about flexibility, 2025 it seems is going to be focused on diversification, both in terms of the services brokerages offer in light of regulatory crack downs and how they approach their infrastructure stacks. This diversification, done well, could open brokerages and trading platforms up to further new opportunities, as well as cost efficiencies and improvements in their infrastructure relationships and set up.
But what do you think? I would love to hear your thoughts on what could be coming in 2025. You can get in touch via my LinkedIn.
Trading expert Mike works with customers not just as a hosting provider but a partner to ensure their infrastructure enables future growth. A golf newbie, he's regularly one shot away from selling his clubs.
