Jinli Technology

Committed to becoming a core switch service provider

Artificial intelligence and high-performance computing

Edge computing and the Internet of Things

Intelligent Network

Data storage and management

Partner Joint Program

CAE (Computer Aided Engineering) solutions

In the industrial and engineering fields, CAE (Computer Aided Engineering) has become an important tool for enterprises to drive product development and updates. High-performance computing platforms serving CAE need to support numerical simulation calculations in various disciplines such as structural analysis, fluid analysis, and electromagnetic analysis, helping enterprises to perform calculations, analysis, research, and simulations of product components, parts, entire vehicles, complete machines, and processes.

Scenario Analysis

In the industrial and engineering fields, CAE (Computer Aided Engineering) has become an important tool for enterprises to drive product development and upgrades. High-performance computing platforms serving CAE need to support numerical simulation calculations in various disciplines such as structural analysis, fluid analysis, and electromagnetic analysis, helping enterprises to perform calculations, analysis, research, and simulations of product components, parts, entire vehicles, complete machines, and processes.

Solution

The high-performance computing platform solution for the CAE field is based on leading-performance, energy-efficient HPC computing power. Through the cluster management and scheduling platform software, it supports customers in conducting engineering simulations in various disciplines such as computational structural mechanics, computational fluid dynamics, computational electromagnetics, and multidisciplinary optimization.

This solution adopts the current mainstream distributed architecture, consisting of three major parts: hardware platform, software platform, and infrastructure.

1. At the infrastructure layer: CPU servers and GPU servers fully support liquid cooling, providing enterprises with green and environmentally friendly infrastructure;

2. At the hardware platform layer: CPU computing servers of various forms and GPU computing servers of various architectures provide ultimate computing performance; industry-leading, feature-rich distributed storage systems meet the requirements of CAE software, especially implicit finite element and electromagnetic analysis software, for file reading performance and data management; computing networks and storage networks of various specifications meet the requirements of CAE software for high-speed data exchange;

3. At the software platform layer: The cluster management and scheduling platform realizes unified scheduling of industrial simulation computing jobs and efficient resource management; the Teye application analysis tool can monitor the computing operation status of CAE application software, analyze performance bottlenecks, and help users optimize performance; the system environment includes operating systems, compilers, parallel program development environments, modes and parallel programming frameworks, and mathematical libraries, etc.

Application Scenarios

This integrated software and hardware solution is widely used in the automotive, energy, electronics, electrical, general machinery, universities, and research institutions industries or fields, providing enterprises with strong computing power and platform software support for engineering simulation business.

Solution Value


Leading Solution Architecture	Flexible Configuration and Easy Expansion	Excellent Computing Performance	Low-Carbon and Environmentally Friendly Product Concept
The platform adopts the industry-leading distributed cluster architecture, connecting industry-leading general rack servers and high-density servers through high-bandwidth, low-latency computing and storage networks, and Inspur's high-performance distributed storage system to provide customers with a powerful computing power infrastructure platform.	The solution adopts the industry-leading distributed cluster architecture, based on InfiniBand Spine-Leaf networking, and can easily expand the cluster's computing power by adding computing nodes; the storage nodes also adopt a distributed architecture, and by adding storage nodes, the storage performance and storage capacity can be improved simultaneously.	The CPU nodes use the new generation of x86 processors to fully utilize the computing performance of the CPU nodes; the GPU acceleration nodes support NVIDIA accelerator cards and support CPU-GPU direct connection design without PCIe Switch, fully releasing the performance of the GPU accelerator cards; the distributed storage system supports NVMe all-flash configuration, providing ultra-high data throughput performance.	Adhering to the green and low-carbon concept, Inspur's CPU computing servers and GPU acceleration servers fully support liquid cooling, and there are also liquid-cooled integrated rack servers and modular liquid-cooled data centers, providing customers with green and energy-saving solutions.