HFT Demands the Very Best Network Performance
The rapid evolution of high-frequency trading (HFT) and algorithmic trading in the financial services market place has only served to increase the emphasis on the provision and, more importantly, assurance of consistently high levels of technical and business performance.
Simply put, the escalation of HFT trading is one of the greater challenges facing network operations and business teams today in financial services firms.
Firms are often forced to dedicate additional and significant resources to monitoring, identifying and resolving any issues that may arise, and into proving best execution. Integration of the specialized low latency monitoring tools with standard network management systems and practices is therefore imperative, especially as these environments grow in scale and business importance. Collecting and reporting on low latency performance metrics along with the supporting HFT and network infrastructure provides a complete and immediate view of all trading operations, ensures best execution services and proof, and reduces operations costs.
You may have read recently in Banking Business Review or MarketsMedia about our partnership with TS-Associates and Terilogy. These partnerships allow us to deliver integrated best-of-breed products that provide Financial Services IT operations teams and business managers with complete and immediate visibility of the performance of their HFT environments, using the same reporting consoles and workflows used to manage their network and servers.
This allows users to monitor critical HFT performance indicators such as hop latency, build baselines of normal utilization levels, and proactively monitor and raise alerts when latency or utilization deviates from historical norms. Once alerted, a user can prevent a potential outage or quickly troubleshoot the network and trading infrastructure.
Also, the traditional approach for performance metric reporting and data retention is to implement a centralized database for historical reporting. This approach is a restrictive and costly practice that results in a significant up front expenditure in both planning and hardware requirements. Further,immediate access to granular data is key and data in its raw format is crucial - aggregated or average metrics rolled up over time exponentially diminish in value.
SevOne addresses the problem of providing complete and immediate reporting with our distributed peer-to-peer architecture, where the real-time collection, storage, and reporting capability is provided with each peer appliance, including the ability to store raw performance metrics for 12 months. This provides significant benefits both in terms of eliminating report generation times, and reducing the capital and operations costs and implementation times of hierarchical reporting systems - the IT operations team does not have to continually administer the system deployment, or overprovision or overbuy capacity for when regulatory retention requirements change. Further, anyone can be provided access to key service assurance reports, from business managers to trading clients.
As firms continue to eradicate unnecessary sources of latency the pressure will build up on the network side – this is where proactive measurement and performance monitoring can help ensure your ultra- low-latency front end is supported by similarly best of breed back end systems.
TS-Associates is at the A-Team Group's Low-Latency Summit 2011 event today in London. The event puts the microscope on low-latency infrastructure that underpins business goals. To read more on the partnership of SevOne and TS-Associates, download the Ensure High Frequency Trading Performance datasheet.
Mike Miracle is the Chief Marketing Officer for SevOne, Inc.