Compute on Demand

Compute On Demand provides the secure and reliable low-latency infrastructure you need to support your trading strategies and meet the demands of your client without the expense and maintenance burden of traditional collocated solutions. 

Compute On Demand is a managed solution optimized for colocated trading environments, built by people who understand the dynamics of the markets. It is available in both physical and virtual editions:

  • The Physical Edition provides a pre-provisioned operating system, hardware network, and storage in a managed solution. It excels in situations where every microsecond counts, because your applications run on your own dedicated server. You retain complete control of what happens when.
  • The Virtual Edition is a financial services cloud community that provides On Demand access to trading infrastructure and services. It gives you the utmost flexibility to quickly add the fully provisioned computing power needed to test your strategies, process large volumes of data or meet the demands of your clients. Operating systems are deployable from within the NYSE Technologies catalogue or Customers can install their own OS by deploying ISOs. Compute On Demand is connected to the SFTI® Network, providing rapid provisioning of access to hundreds of buy-side and sell-side traders around the world, as well as historical and real-time information, and services from both NYSE Technologies and third-party vendors. New services can typically be added in a matter of days in the physical edition, and minutes in the virtual edition.



Business Challenges Compute On Demand Solutions

Reducing time to market

It can take a long time to work through your firm’s procurement systems to get equipment purchased and installed. Adding new services can then take weeks.

Rapid start-up and expansion

All equipment is pre-provisioned, reducing startup times from several months to a couple of weeks. New SFTI services can be added in days.

Cutting the cost of a low-latency infrastructure

Building your own low-latency trading infrastructure requires large capital investments.

No capital expense required

With Compute On Demand, NYSE Technologies owns the infrastructure. You simply access as much of it as you need, either virtually or by leasing a server.

Adapting to rapidly changes needs

Building and maintaining your own infrastructure requires long-term commitments to a set amount of capacity. Adding, or subtracting, servers is neither cheap nor easy.

Highly scalable solutions

Both the physical and virtual editions offer low-cost base packages that can be easily and quickly ramped up to meet increased demand. Additional storage is available from via a fiber attached EMC SAN.

Keeping hardware and software up to date

It takes a lot of time and technical expertise to research and build out low-latency trading environment. Keeping that infrastructure current may then involve negotiating a lengthy internal procurement process.

State-of –the-art equipment

NYSE Technologies takes full responsibility for building, upgrading and maintaining a state-of-the-art trading infrastructure.

Retaining the expertise to maintain infrastructure

A large and dedicated operational group is required to maintain and monitor all aspects of trading solution, at all hours of the day and night.

24/7 maintenance and monitoring

NYSE provides management and monitoring of the hardware, network, operating system stack, and storage backed by our 24x7 Service Desk. Our market data and order entry expertise provides unrivaled support.

Maintaining a secure platform

Making sure that your infrastructure supports a safe and secure trading platform requires a significant expertise and constant vigilance.

World Class Security

NYSE’s security team ensures the SFTI network is protected against threats. The servers are updated with the latest security

Competitive trading demands an infrastructure capable of handling the complexities of today’s markets. Compute on Demand will give you the means to manage them today, and tomorrow.