Computing
Three major concerns of industrial CAE users are accuracy, computing time and user-friendliness. In modern simulation, since the geometrical of design phase is getting more complicated day by day, true 3D model and simulation is in great demand. However, true 3D simulation inevitably increases computing time and requires more memory. The demand for high accuracy and high performance computing can never be underestimated. Generally, latest and more powerful CPU can reduce computing time. However, this type of improvement from CPU clock rate alone cannot satisfy industrial users neither in speed nor accuracy.
Utilization of multiple multi-core CPUs is therefore the most effective approach. Hence, parallel computing is fast becoming an inexpensive alternative to standard supercomputer for solving large-scale problems that arise in scientific and engineering applications. Parallel Processing (PP) can extract the full power of your computer. But if you need more, Moldex3D provide remote HPC function which use internal servo for simulation. We also provide Cloud-Connect to access cloud computing provider like Amazon (AWS), Microsoft (Azure) and Google Cloud Platform (GCP) etc., which give flexible and powerful computing resource.
Moldex3D also support Linux, an open-source operating system known for its high performance, stability, security, and customizability. It offers a wide range of software and applications, has a strong community support, and is cost-effective. Give an alternative platform for Moldex3D simulation.
Capabilities
- Parallel Processing (PP)
- Remote computing
- Support Microsoft Windows and Linux platform
- Cloud-Connect
Features
Parallel Processing (PP)
- SPMD – Single Program Multiple Data, each processor executes the same program but applies different data
- Single Node Computing for a multi-core CPU computer
- Cluster Computing for a group of multi-core computers connected by network
- Moldex3D uses METIS, which developed by Karypis Lab of Minnesota University, as the mesh partition kernel.
Remote Computing
- Computing Manager and Job Scheduler to help users and administrators manage and run job analyses
- Computing manager is the tool for scheduling project analyses and support cross-project analyses.
- Job Scheduler is for administrators to manage accounts, compute nodes, jobs submitted to the Job Scheduler Service, and to change the settings.
- Moldex3D also support PBS Pro, SLURM.
Linux Platform
- Linux is a high performance, high reliable, stable and open source operating system
- Moldex3D 2023 support CentOS 7 series, CentOS 8 series, RHEL 7 series, RHEL 8 series.
Cloud-Connect
- Toolkit to help user to build the cloud computing environment quickly and automatically.
- Easily deploy self-computing farms on Amazon AWS, Microsoft Azure and Google GCP with own account.
- Flexible select compute node with different specifications
- AutoScaling function can automatic increase/decrease cloud compute nodes base on the job queue status and auto-start / stop. No waste and extra charge