Tech-X engineers and scientists have extensive experience in developing highly performant code.
We can improve the performance of your software, systems, and workflow. We have specialized expertise in the following areas:
- GPU Computing (CUDA, OpenCL)
- Cluster computing (MPI and distributed memory)
- Linear algebra, including the use of High-performance libraries, such as PETSc and Trilinos
- Advanced parallel instructions (e.g., x86 AVX)
Using techniques based on high-level programming languages such as C++, fourth-generation languages like IDL and MATLAB, popular scripting languages such as Python, advanced visualization tools, and newer architectures like Graphics Processing Units (GPUs), we develop workflows and solutions to assist in your data discovery process.
Tech-X can improve the efficiency of your computational solutions based on our hardware knowledge and algorithm implementation. We have extensive knowledge of computational kernels such as Krylov methods, FDTD, ray tracing, finite volume including AMR, high order finite elements, and method of moments. We also have demonstrated success in leveraging existing libraries such as BLAS, ScaLAPACK, FFTW, Trilinos, and the Portable, Extensible Toolkit for Scientific Computation (PETSc).
What Tech-X Customers Say
“My experience of Tech-X can best be described as amazement after amazement. Before we decided to seek out their help, I very much doubted how quickly they could grasp the essence of our complicated algorithm, let alone drastically speeding it up. They immediately proved my concern was unnecessary and produced a detailed report, free of charge. Within the time promised, the converted GPU-based code was delivered. I was impressed by the overall speed improvement, and in particular amazed by the speedup (~80x) in one part of the code that I had thought was too computationally involved to be GPU-worthy. In Phase 2 of the project, further speedup was delivered as promised... the code was also very nicely refactored to the degree of almost an overhaul... a valued bonus because I had been... afraid to touch the complicated code base. Overall I enjoyed working with Tech-X and strongly recommend handing them your computational projects.” – Dr. Lin Shao, Betzig Lab, Howard Hughes Medical Institute
“Tech-X provided me with an interface to the PETSc library built to my specifications. Using this new solver accelerated my code by a factor of 80.” – Dr. Henry Freund, SAIC
“GPUs are a game-changer for simulating antenna behavior on electrically large platforms and Tech-X is helping us take full advantage of them.” – Dr. John S. Asvestas, Naval Air Systems Command (NAVAIR)
“GPULib caught my eye immediately as a tailor-made way of boosting the speed of my remote sensing algorithms without having to write my own CUDA kernels.” – Dr. Mort Canty, Forschungszentrum Juelich, Germany
“Tech-X’s innovative tools are instrumental for visualizing our complex astrophysical simulations and observations.” – Dr. Daniel Pomarede, CEA, France
“Tech-X's expertise in both numerical methods and GPU computing make them ideal partners for porting complex simulation codes to GPUs.” – Dr. Dave Swensen, Reaction Engineering Intl., Salt Lake City
For more information about how Tech-X can improve your software, systems, and workflow, contact Tech-X Sales.