The Software Built to Perform

The award-winning Xcelerit SDK is a software toolkit to boost the performance of compute-intensive applications while preserving programmers’ productivity. From a modified sequential source code – free from any parallel constructs, low-level compiler directives, etc. – the Xcelerit SDK can generate highly optimised code for a variety of target processors. These include GPUs, many-core processors – and combinations of these in a grid. The programming interface is intuitive and suitable for many classes of algorithms, e.g. Monte-Carlo, Finite-difference, or Lattice-based methods. The Xcelerit SDK is targeted at domain specialists performing heavy computations with the need for a flexible in-house solution. Users can focus on the core of their algorithm and the problem at hand, and yet they can generate high-performance programs.

Xcelerit SDK Box


Dramatically boost the performance of existing applications


Easily develop high-performance applications

Cost Efficiency

Massively reduce development and infrastructure costs

Supported Hardware

  • GPUs
    GPUs from Nvidia, Fermi, Kepler, Maxwell, or Pascal architectures
  • CPUs
    x86 CPUs from Intel or AMD
  • Intel Xeon Phi

    Intel Xeon Phi accelerator processor (KNC)

Supported Operating Systems

Supported Extensions

  • Matlab

    Interface with Matlab from Xcelerit-enabled code

  • Microsoft Excel

    Interface with Excel from Xcelerit-enabled code

  • IBM Platform Symphony

    Seamlessly deploy Xcelerit-enabled applications on your IBM Platform Symphony grid

Mike Giles - Professor of Mathematical Finance and Scientific Computing

I really like the Xcelerit SDK for Monte Carlo applications. The API is very clean, and the implementation is very elegant. I can see this approach doing very well with the banks who will like this better than the turn-key solutions that most other ISVs develop.

Mike Giles - Professor of Mathematical Finance and Scientific Computing / University of Oxford

Credit Risk Quant

Xcelerit technology helped us boost a complex credit risk computation. No expert programming skills were required and with only minor changes to our original sequential code we observed a dramatic performance increase.

Credit Risk Quant / Societe Generale

Selected Resources

Chief Architect /

IBM Platform Computing really does a good job of controlling exactly what resources are made available to any one job – the Xcelerit Software Development Kit then makes sure that every available chunk of compute power is used to the limit.

Chief Architect / IBM Platform Computing

Principal Engineer /

Intel processors can really deliver spectacular performance to programmers who make good use of their cores, caches and vector processing units. Xcelerit technology opens up Intel performance to programmers who don’t have that expertise.

Principal Engineer / Intel

Business Development Manager /

NVIDIA GPUs can deliver staggering power to the end user. In many financial applications, we find that companies have quite extensive existing codebases. Xcelerit makes it very easy, with minimal modifications, to enable really quite spectacular speed-ups.

Business Development Manager / Nvidia