The London Stock Exchange Group (LSEG) and Nvidia are working to reduce the time required for financial institutions to analyse vast quantities of historical market data, reflecting a broader shift towards faster, more data-intensive decision-making.
In a recent announcement, LSEG said its collaboration with Nvidia would use GPU-accelerated computing to process its Tick History datasets more quickly. According to the companies, benchmark analysis using Nvidia’s DF technology and Polars reduced processing times for years of historical market data from hours to seconds. The performance claim relates to their benchmark analysis and should not be interpreted as a guarantee that every financial workload will achieve the same result.
The scale of the underlying dataset helps explain the significance of the development. LSEG says Tick History contains approximately 110 trillion rows covering 100 million financial instruments, more than 30 years of historical information and data from over 580 exchanges, trading venues and contributors.
The companies say the collaboration is intended to make this information easier to analyse without requiring customers to create another full copy of the underlying dataset. It is also designed to preserve the licences, permissions and controls governing access to the information.
From historical data to faster decisions
Financial institutions use historical market data to investigate price movements, test trading strategies, reconstruct market conditions, evaluate execution quality and examine risk. These activities can involve processing large volumes of records before analysts can begin interpreting the results.
Reducing processing time could allow researchers and risk teams to examine more historical information and test a wider range of scenarios. It may also help institutions move beyond workflows that depend on overnight data processing, particularly as markets evolve towards longer trading hours.
LSEG reported that traffic on its real-time network grew by 29% in the first half of 2026 compared with the same period a year earlier, reaching a peak of 26 million messages per second in June. Customer API requests for Tick History data increased by 48% over the same period, according to the company. These figures suggest growing demand for timely access to market information, although they do not independently measure the commercial impact of the Nvidia collaboration.
The partnership also highlights the importance of data governance. Financial institutions cannot simply make every dataset available to every model or analyst. Licensing restrictions, access permissions, confidentiality and regulatory obligations remain relevant even when processing becomes faster.
Implications beyond financial services
The wider lesson applies to companies holding substantial volumes of customer, operational, transaction or supply-chain data. The value of information depends not only on how much an organisation possesses, but also on how quickly it can turn that information into a decision.
AI investment is frequently discussed in terms of model capability. Yet the infrastructure connecting data to analytical tools can be just as important. Faster processing may create opportunities, but organisations must still ensure that their information is accurate, relevant and properly governed.
For African companies, this does not mean that every business needs advanced GPU infrastructure. It means that data architecture and accessibility should be treated as strategic considerations rather than back-office technical issues.
BrandiQ Takeaway: The next stage of enterprise AI will be shaped not only by smarter models but by the speed, quality and governance of the data behind them. Competitive advantage belongs to organisations that can convert trusted information into useful decisions without losing control of it.
