The Race Below the Nanosecond
Why Sub-Nanosecond Timing Is Now a Financial and Regulatory Necessity
In December 2025, the Wall Street Journal reported on a dispute between Mosaic Finance and Eurex, one of the world's largest derivatives exchanges. The complaint centred on a single technical manoeuvre that French trading firm Mosaic calls "corrupted speculative triggering." Its effect: shaving roughly 3.2 nanoseconds off reaction time to market-moving data.
3.2 billionths of a second. According to Mosaic, a group of high-frequency trading firms has used that edge to generate hundreds of millions of dollars in profit.
This was not an isolated curiosity. It was confirmation of something the market has known for years: time is not a backdrop to trading. It is the trade.
And it raises a question every CTO, risk officer, and head of infrastructure in financial services should be asking right now. If 3.2 nanoseconds is worth hundreds of millions of dollars, what does your organisation's timing infrastructure actually deliver and can you prove it?
The Money Is Real. So Is the Exposure.
High-frequency trading firms exploit fleeting price differences between related instruments, a Euro Stoxx 50 move that should, a fraction of a second later, ripple into DAX futures. Whoever reacts first captures the spread. Whoever reacts second pays for it.
This dynamic has been building for over a decade. Off-the-shelf systems already timestamp to within 100 nanoseconds using GPS-disciplined clocks. Modern time synchronisation systems provide resolution down to one nanosecond for automated trading activity purposes, as reflected in the FIX Performance Session Layer (FIXP) standard and the FIX Time Precision Technical Addendum. Regulators in the EU have, since 2015, pushed exchanges toward nanosecond-level clock synchronisation specifically to police this kind of advantage.
The Eurex dispute shows where that race has now arrived: fractions of a nanosecond, contested in public, with regulators watching and hundreds of millions of dollars at stake.
But speed is only one half of the equation institutions now face. The other half is proof and it is just as commercially consequential.
A Regulatory Landscape That Has Quietly Become Enormous
Financial institutions operating across borders are not dealing with one timing requirement. They are dealing with a dense, overlapping web of obligations spanning operational resilience, cybersecurity, data protection, and third-party supply chain risk - and for large, internationally active banks, that landscape can run into the hundreds of distinct regulatory requirements across the jurisdictions in which they operate.
In the UK and EU, the Digital Operational Resilience Act (DORA) requires financial entities and their critical technology suppliers to demonstrate resilience against ICT disruption, with fines of up to 2% of global annual revenue for non-compliance. In the UK specifically, the Bank of England, the PRA, and the FCA now require firms and financial market infrastructures to report operational incidents and register material third-party arrangements, including the infrastructure that keeps their systems synchronised.
In the United States, Executive Order 13905 directs federal agencies and critical infrastructure operators to build resilience against disruption to GPS and other single sources of positioning, navigation, and timing. The Department of Homeland Security has since published detailed best-practice guidance for resilient PNT in critical infrastructure, explicitly naming financial systems as a sector that cannot afford to depend on a single timing source.
In the UK, the government has gone further still. In November 2025, the Department for Science, Innovation and Technology committed £155 million to national timing resilience, including £68 million for the National Physical Laboratory's National Timing Centre - the UK's first nationally distributed timing infrastructure, built specifically so that financial trading platforms, telecoms networks, and other critical services are not dependent on a single satellite source for time.
This is the second half of the sub-nanosecond story. It is not only about who can react fastest. It is about who can demonstrate, to regulators, auditors, and counterparties, exactly when something happened, on what authority, and with what resilience against failure.
Two Pressures, One Infrastructure Decision
Put these two forces together and the picture sharpens considerably.
Financial institutions need more accurate time, because speed is now directly monetisable down to single-digit nanoseconds, as the Eurex dispute makes clear. And financial institutions need provable, resilient, UTC-traceable time, because regulators across the UK, EU, and US increasingly treat timing infrastructure as a critical dependency in its own right - not a background utility, but a component of operational resilience that must be assessed, registered, and defended like any other.
These pressures used to be considered separately: a trading desk problem and a compliance problem, owned by different teams, solved with different budgets. That separation no longer holds. The infrastructure that delivers a competitive timing edge is the same infrastructure regulators expect institutions to be able to prove, audit, and recover.
This is why an increasing number of institutions are evaluating what we describe as a sub-nanosecond multi-layered timing ecosystem: an architecture that delivers resolution finer than the standard nanosecond clock, while remaining UTC-traceable, auditable, and resilient against GNSS jamming, spoofing, and single-source failure.
Done well, this is not a trade-off between speed and compliance. It is the same investment, serving both.
Why Existing Approaches Fall Short
The two most precise commercially available timing technologies today are White Rabbit, originally developed for CERN's particle accelerator, and SyncE (Synchronous Ethernet).
White Rabbit delivers genuinely impressive point-to-point accuracy, into the picosecond range under controlled conditions. But it demands specialist hardware throughout the network: a White Rabbit-enabled card in every device, a White Rabbit-compatible switch at every hop and a dedicated timing-only network to carry the timing signal. For an institution running thousands of servers across multiple data centres and cloud regions, that is not an upgrade. It is a wholesale infrastructure replacement - expensive, slow, and tied to a small number of hardware vendors.
SyncE, meanwhile, synchronises clock frequencies with sub-nanosecond precision but cannot timestamp events at that resolution. It keeps clocks in step. It does not produce the auditable record that regulators and risk teams actually need.
Neither approach solves the problem at the layer that matters most: the application layer, on the standard server infrastructure that institutions are already running, distributed across on-premises data centres and multiple cloud providers.
The Hoptroff Approach: Software-Defined, Supply Chain Agnostic
This is where Hoptroff's position is distinct, and deliberately so.
Hoptroff is developing the capability to deliver sub-nanosecond timing resolution and accuracy through software, via the Hoptroff TSE® boundary clock and enterprise client, without requiring institutions to replace their existing network hardware.
The architecture uses White Rabbit for time delivery across distances previously considered beyond its practical reach, to the client's point of demarcation at nanosecond-level accuracy, UTC-traceable back to national and international reference standards. From there, Hoptroff's software layer extends that precision through to the application level, on the infrastructure institutions already operate.
Critically, Hoptroff is hardware and vendor agnostic. Whether an institution's estate runs on Safran, Microchip or other timing hardware, on-premises racks or multi-cloud deployments, Hoptroff's layer sits above the physical layer rather than inside it. That matters more than it might first appear: under frameworks like DORA, third-party concentration risk is now a regulatory concern in its own right. A timing strategy locked to a single hardware vendor is itself a supply chain vulnerability. A software-defined, vendor-agnostic layer is not.
The result is an institution that can pursue sub-nanosecond performance for its trading and data infrastructure, while simultaneously strengthening the auditable, resilient timing posture that regulators in the UK, EU, and US are now actively scrutinising.
Modernising Timing for What Comes Next
None of the pressures described here are temporary. Transaction volumes will not fall. The contest for nanosecond and sub-nanosecond advantage will not slow, as the Eurex dispute demonstrates. And this will not be the last timing-based trading technique that puts all traders at risk. Regulatory expectations around operational resilience, third-party risk, and cybersecurity will continue to expand, not contract, across every jurisdiction in which institutions operate.
At the same time, AI-driven trading and risk systems are raising the bar on timing resolution further still, since they react to market and operational data in real time and require the finest possible temporal granularity to function reliably. Distributed databases, increasingly spread across hybrid on-premises and multi-cloud environments, depend on precise time agreement across every node simply to commit data consistently - the finer the synchronisation, the better the system performs.
This is the strategic case for modernising timing infrastructure now, with Hoptroff. Not as a narrow upgrade to meet today's regulatory floor, but as a foundational investment that improves trading and operational performance, prepares infrastructure for AI-driven workloads, and supports institutions' hybrid strategies across on-premises and cloud, without tying that investment to a single hardware vendor's roadmap.
The institutions treating timing as infrastructure, rather than background utility, are the ones positioning themselves to compete on speed and withstand scrutiny on proof. Both, at once, on the same investment.
Hoptroff delivers UTC-traceable, sub-nanosecond precision timing for financial services, government, energy, transportation and critical infrastructure - through software, without wholesale hardware replacement, and without dependency on a single vendor's supply chain. Traceable Time as a Service (TTaaS®). Time you can trust, prove, and operate on.