Posted inOpinion

Financial institutions are unwittingly extending their attack surface due to a firewall fixation

In an industry where trust and risk reduction are paramount, shrinking unnecessary threat surfaces should be a top priority.

Cyber security, Cyberattack, hacker
Cybersecurity. Credit: Shutterstock

In the financial sector, technology is the pulse that sustains the business. Modern banks, investment firms, and trading platforms rely on a complex web of IT systems to facilitate everything from split-second trading decisions to seamless customer experiences. Think of the billions of data packets flying every minute — account balances updated across apps and ATMs, real-time market data streamed from Bloomberg or Reuters to traders, or high-frequency transactions executed in milliseconds. These networks are vast, interconnected, and under relentless pressure to perform flawlessly.

Yet with this complexity comes exposure. Every additional connection, each new data flow, potentially opens a door to attackers. And the finance sector, where data is power and trust is currency, remains a prime target for cybercriminals.

The growing threat to financial institutions

The evidence is clear. The IBM Cost of a Data Breach Report 2024 found that the financial sector suffered the second-highest average breach costs of any industry, at $5.9 million per incident. In a region like MENA, where financial services are growing rapidly and embracing digital transformation at pace, these risks are magnified. Institutions are under pressure to innovate quickly. But every innovation adds complexity to their security posture.

In such a scenario, financial institutions deploy a multi-layered security strategy that includes endpoint protection, intrusion detection, encryption, zero-trust frameworks, and, of course, firewalls.

Firewalls have been the foundation of IT security for decades. Their strengths are well known. They act as gatekeepers, inspecting and filtering incoming and outgoing traffic based on pre-set rules. They are flexible, widely understood by IT teams, and scalable. It’s no surprise they are a default component in almost every financial institution’s perimeter defence.

But herein lies the problem. Because they are so familiar and because they have largely stood the test of time, many organisations lean too heavily on them. The thinking seems to be: if it isn’t broken, why fix it? Unfortunately, it may well be broken, or at the very least, badly frayed.

Why firewalls alone fall short

Overreliance on firewalls creates a significant blind spot. Firewalls were not designed to meet the sophisticated threats financial services now face. Attackers have developed a host of techniques that evade or even exploit firewalls, such as application-layer attacks, encrypted malware, and insider threats.

Firewalls are software-based, which makes them susceptible to misconfiguration. This is a common problem in complex environments. They also operate bi-directionally by default. That’s fine when you trust both sides of the connection, but dangerous when an attacker has already breached one side and is using the firewall to exfiltrate sensitive data.

Ultimately, a firewall’s job is not to guarantee data integrity or enforce strict one-way flows. It’s simply to filter according to rules. And those rules can be bypassed.

Physical protection in a digital world

Decades of digital transformation have left many believing that everything can be solved digitally. But just as banks still need both physical branches and mobile apps to fully serve their customers, protecting critical networks requires more than just digital tools. There are physical elements of infrastructure that are equally, if not more, important in keeping data safe.

This is where data diodes come in.

A data diode is a hardware-based security device that allows data to flow in only one direction. Unlike a firewall, which relies on software rules and configurations, a diode is a physical mechanism. By immutable design, they make it impossible for data to flow back. There is no reverse channel to exploit, no configuration error to undo years of planning.

For many financial professionals, data diodes remain something of an unknown entity. That unfamiliarity, combined with concerns about latency in a world where milliseconds count, are key reasons they’ve largely flown under the radar. Yet modern diodes, such as those from OPSWAT, can achieve throughput of up to 10 gigabits per second, making them perfectly suited to high-speed trading and real-time operations.

Of course, it’s worth noting that because data diodes enforce strict one-way communication, they are not appropriate for every scenario. But for several, high-risk use cases, they are unmatched in their ability to protect sensitive systems while still enabling essential data flows.

Where data diodes shine

So where should financial institutions consider deploying data diodes?

Take backup and archiving of sensitive data. Financial institutions routinely back up operational data to secure archives to safeguard against system failures. A diode can transfer files and replicate databases into the archive without creating a bidirectional link that could expose the operational network.

Or consider the secure transfer of real-time market data. Trading floors rely on feeds from Reuters and Bloomberg that are inherently one-way: the data flows into the trading environment but doesn’t need to flow back out. A data diode at the network boundary ensures those feeds arrive securely and quickly, without opening the door to attackers.

Regulatory reporting is another strong use case. Banks and insurers regularly send compliance reports to regulators from secure environments. This is a one-way push of highly sensitive data, ideal for a data diode, which ensures the data leaves securely without compromising the source network.

Fraud detection and transaction monitoring systems also benefit from strict network segmentation. Logs and transaction data can be streamed into isolated detection systems via a data diode, allowing real-time anomaly detection without exposing core banking systems.

For institutions that rely on SPLUNK for security and operational analytics, data diodes offer a secure way to export log data from sensitive environments to SPLUNK, maintaining strict separation, while supporting real-time visibility.

And, of course, many financial firms are migrating workloads and data to the cloud. Using a data diode to replicate sensitive data to cloud platforms ensures the security of internal networks remains intact.

Rethinking risk reduction

Firewalls still have an important place in modern IT security. But viewing them as the ultimate line of defence is a restricted way to think about risk management. In an industry where trust and risk reduction are paramount, shrinking unnecessary threat surfaces should be a top priority.

Data diodes are not a hypothetical solution. They are already used extensively in critical national infrastructure, where failure or compromise could put lives at risk. Originally designed for military and defence systems, they are a well-proven technology that brings the same level of assurance to financial environments.

The reality is simple: in a world where attackers are getting smarter, defenders cannot afford to stand still. Adding data diodes to the mix to create a layered approach isn’t just good practice. It’s the kind of risk mitigation the industry owes itself and its customers.