We are customer-centric and provide customized or one-stop full-stack solutions to empower all industries.
CAE, or computer-aided engineering design, is an engineering analysis technology based on modern computational mechanics and computer simulation computing power. It is the main support module for achieving product optimization design. Currently, in the aviation, aerospace, energy and power industries, using CAE for repeated design, analysis, and optimization has become a standard necessary step and means. Different CAE applications have different requirements for hardware resources such as processors, networks, and storage. In response to the needs and challenges of CAE high-performance computing platforms, PowerLeader has proposed the main solution for CAE high performance.
At present, in the aviation, aerospace, energy and power industries, using CAE to repeatedly design, analyze and optimize on HPC clusters has become a standard and necessary step and means. Different CAE applications have different requirements for hardware resources such as processors, networks and storage, and require a lot of resources to select, deploy and maintain. In order to solve the problems of difficult deployment, selection and maintenance in the CAE field, Powerleader has launched an HPC solution for CAE, which simplifies the deployment of HPC clusters while providing high flexibility, scalability and high performance, boosting the development of the CAE field.
In terms of hardware, the computing nodes of PowerLeader CAE high-performance computing solution are interconnected through a standard open high-speed network, running an open source Linux system, and providing a single system interface to the outside world. It is designed for large-scale scientific parallel computing. It has both transaction processing and network information services.
In terms of software, the PowerLeader cluster system is equipped with a rich software system, including operating systems, compilers, debuggers, function libraries, parallel communication libraries, global shared parallel file systems and other basic software.
PowerLeader PR2715W is a 2U server product based on the third-generation Intel® Xeon® Scalable Processor. It uses the Intel C620 series high-performance chipset and supports up to two third-generation Intel® Xeon® Scalable Processors. At the same time, PCIE is upgraded to 4.0, which greatly improves the communication capabilities between the CPU and PCIE devices. The overall performance of PR2715W is improved by up to 40% compared with the previous generation of products, with a higher processor TDP, stronger hardware security capabilities, and better energy efficiency.
The GPU node uses PowerLeader PR4910W, which is the latest 4U dual-socket rack-mounted accelerated computing server based on the third-generation Intel® Xeon® Scalable processor launched by PowerLeader. This model has 32 DDR4 DIMM slots and a maximum memory capacity of 8TB; it has 10 full-height, full-length, double-width GPU slots, and provides 2 PCIe 4.0 x16 slots and 1 OCP network card slot. It provides support for the acceleration of application software and realizes a rapid improvement in CAE efficiency.
In order to cope with the characteristics of most CAE software with good parallel scalability and large amount of calculation data, PowerLeader high-performance cluster is equipped with PL-XDFS distributed parallel storage system. PL-XDFS launched by PowerLeader supports massive data, online expansion, and synchronous improvement of storage capacity and aggregate bandwidth, meeting the requirements of high-throughput computing for storage systems.
The cluster is equipped with a high-bandwidth, low-latency Infiniband high-speed network, which meets the data exchange requirements of various MPI programs in application software, improves the parallel efficiency between multiple nodes, and speeds up the simulation calculation process.
PowerCloud cluster management and job scheduling system realizes a unified portal for design and simulation platforms, integrated management, monitoring management and job scheduling.
Supports Fortran77/Fortran95/C/C++, etc., and supports OpenMP and MPI program compilation. In addition, we also provide parallel program debugging and tuning tools to help users better complete program development. PowerLeader high-performance cluster system is equipped with a rich mathematical function library and parallel environment for users to choose from, including Goto, Lapack, FFTW, MPICH, etc.
PowerLeader CAE high-performance computing solution is a cluster architecture, using PowerLeader’s new generation server PR2715W built on the third-generation Intel® Xeon® Scalable processor as the computing node, and PR4010W as the GPU acceleration platform. All hardware and software are deployed in the system to achieve efficient parallelization, collaborative simulation, and data security sharing integration.
PowerLeader’s high-performance cluster servers all use the X86-64 platform and are homogeneous systems that can share software resources and data storage space. Resource scheduling is simple, and node ratios can be flexibly adjusted to meet various application requirements, effectively reducing users’ total cost of ownership.
The computing resources are rich, meeting various requirements of CAE applications and providing a suitable environment. Intel cryptographic operation hardware acceleration technology is also used to reduce the impact of encryption calculations on performance and ensure efficient use of resources.
This solution uses high-bandwidth, low-latency InfiniBand as a communication network to improve the communication capabilities between nodes, greatly reducing communication latency. And all computing nodes are fully interconnected through IB switches.
The entire cluster adopts a variety of reliability design solutions to greatly improve the reliability of the system and ensure that there is no single point of failure in the system.
Ability to flexibly expand existing computing resources and storage space according to user needs.
