Witness the impressive performance of CONVERGE CFD software
running on a Nor-Tech optimized server integrated with Intel Xeon processors
Converge CFD Benchmark
We just completed a benchmark using Intel and MPICH mpi that are integrated in with Convergint’s Converge 3.1.6. 32 cores on 1 node vs 64cores on 2 nodes. The interconnect used was InfiniBand 100GB Direct connect. Pretty impressive! See the benchmark graphic attached here.
CONVERGE CFD Software
Convergent Science, headquartered in Madison, Wisc. with offices and distributors around the world, is an innovative, rapidly expanding computational fluid dynamics (CFD) company. Their flagship product, CONVERGE CFD software, removes the grid generation bottleneck from the simulation process.
CONVERGE is a cutting-edge CFD software package that eliminates all user meshing time through fully autonomous meshing. CONVERGE automatically generates a perfectly orthogonal, structured grid at runtime based on simple, user-defined grid control parameters.
With fully coupled automated meshing and Adaptive Mesh Refinement (AMR), CONVERGE can easily analyze complex geometries, including those with moving boundaries. Moreover, CONVERGE contains an efficient detailed chemistry solver, an extensive set of physical sub-models, a genetic algorithm optimization module, and fully automated parallelization.
CONVERGE is used by companies, academic institutions, and national laboratories around the globe to simulate everything from engines to battery packs to wind turbines.
3rd Gen Intel® Xeon® Scalable Processors
The 3rd Gen Intel Xeon Scalable processors, with AVX-512, are optimized for a range of simulation applications, all with the consistent, compatible, Intel architecture you have come to depend on.
Highlights Include:
- Up to 40 cores/socket
- Multi-socket support (1,2 CPU)
- Up to 3 UPI channels per CPU
- Validated for Intel 3D NAND SSDs
- PCI Express 4 and 64 lanes (per socket) at 16 GT/s
- Support for up to 3200 MT/s DIMMs (2 DPC)
- 16 GB‒based DDR4 DIMM support, up to 256 GB DDR4 DIMM support
- Select SKUs will support a maximum memory capacity of 6 TB/socket
Innovations in core architecture and memory bandwidth, combined with Intel’s oneAPI HPC Toolkit to analyze and optimize applications for Xeon platforms, ensure developers can expect fast, accurate and stable performance on Intel based systems
Nor-Tech Integration Expertise
Putting it all together, Nor-Tech has one of the most highly respected technology engineering teams in the world—capable of everything from minor tweaks that solve big problems to complex hardware and software integrations that other engineers won’t touch. Linux expertise is one of the many areas where we excel.
Application Integration
With most hardware proposals, integration expertise is a deciding factor; the fact is, many technology companies can sell the hardware, but very few can leverage it for maximum productivity. Integrating applications in a distributed parallel environment is extremely complex and requires a collaborative team of experts. Most of our engineers have been with us for a decade or more and work closely with industry partners such as Convergent Science.
Hardware Integration
We source hardware from the industry’s leading developers, such as Intel; working closely with these partners to engineer an optimal configuration that is reliable, powerful, scalable and cost-effective
Fast, Accessible Support
The same Nor-Tech engineers that build your technology are readily accessible during business hours to answer questions and explain complicated procedures in terms that even those without a technology background can understand. Our engineers get to know our customers—there’s always a friendly, familiar voice on the other end of the phone line.
Intel, Convergent Science and Nor-Tech
Collaborating for top performing, cost-effective simulation solutions.
To schedule a free trial
Sign up at: https://www.nor-tech.com/solutions/hpc/demo-cluster/
or call 952-808-1000