NX Nastran 4 Extends Capabilities in High-End CAE
With the release of NX™ Nastran 4, UGS has very strong tools that continue to grow its solver workhorse behind the I-deas® NX Series, FEMAP® and NX™ software product lines. Two major thrusts for this latest release are the enablement of 64-bit processing (particularly on 32-node and 64-node clusters of inexpensive 64-bit Xeon dual processor servers) and overall NX Nastran product enhancements.
Leading performance for 64-bit technology and parallel processing
All modules of NX Nastran can now fully exploit the memory advantages of 64-bit architecture. UGS’ work to take full advantage of all benefits from both a software and a hardware angle continue. Paresh Pattani, director of HPC and Workstation Applications at Intel notes, “Intel has been working closely with UGS applications for over eight years to improve performance and optimization and is now delighted to help deliver superior NX Nastran 4 performance on both the 64-bit Intel® Xeon™ and Intel® Itanium® 2 platforms...”
Customer-driven enhancements
Based on customer input, NX Nastran 4 includes a variety of major enhancements in three primary categories.
Performance improvements
- Extensions to Hierarchic Distributed Memory Parallel (HDMP) solver – Improvements in distributed parallel solutions have greatly increased the performance for large complex models and now support dynamic response analyses. NX Nastran can solve a very large product performance simulation, leveraging parallel processing with up to 64 computers. This can translate to major productivity improvements compared to competitive solvers.
- Memory management improvement – NX Nastran 4 memory handling has been re-architected, dramatically increasing the amount of RAM that can be accessed to more than a million terabytes. Memory needs are now only limited by hardware systems. This will significantly improve the overall performance of large-solve scenarios and make possible solutions of large models that could not be performed previously.
- Lanczos Eigensolver performance improvement – Improvements to the standard Lanczos Eigensolver can enable 15-25% reductions in solution times for large shell element dominant models.
- HP-MPI technology is leveraged by NX Nastran across all applicable platforms, as a way of increasing scalable performance, robustness, and platform flexibility.
Discipline extensions adding capabilities such as
- Extensions to advanced non-linear solver by including more hyperelastic materials models, improving solution algorithms, and adding more non-linear element models.
- Explicit non-linear solver capability for efficient simulation of high-speed impact events and metal forming analysis is now available.
- Rotor Dynamics capability that enables engineers to develop designs that operate away from the critical unstable operating speeds of rotating systems
- Linear contact capability allows users to perform fast and efficient surface-to-surface contact solutions for simulations that are otherwise linear.
- Curved Beam Element Enhancement supports updated ASME boiler and pressure Bessel code requirements for the design of flanged and mitered pipes.
- Modal Energy Method provides engineers with greater insight to understanding the dynamic behavior of their systems.
Process improvements
- Enforced motion improvements – This extended capability for dynamic enforced motion simulation improves accuracy.
- Temperature dependent modeling of composites – The latest enhancements support more accurate calculations of thermal loads on composite structures
As the solver backbone for UGS’ CAE Solutions, NX Nastran is an important part of an overarching digital simulation strategy to enable users to implement their own best practices and increase simulation’s influence on design.
