Expand/Collapse all
1. Overview of Parallel Processing
1.1. Specifying Processes, Threads, and Cores in SMP, DMP, and Hybrid Parallel
1.1.1. Auto-hybrid
1.2. Parallel Processing Terminology
1.2.1. Hardware Terminology
1.2.2. Software Terminology
1.3. HPC Licensing
2. Using Shared-Memory Parallel (SMP) Processing
2.1. Activating Parallel Processing in a Shared-Memory Architecture
2.1.1. System-Specific Considerations
2.2. Troubleshooting
3. GPU Accelerator Capability
3.1. Activating the GPU Accelerator Capability
3.2. Supported GPU Hardware, Analysis Types, and Features
3.2.1. GPU Hardware
3.2.2. Supported Analysis Types
3.2.3. Supported Features
3.3. Troubleshooting
4. Using Distributed-Memory Parallel (DMP) Processing
4.1. Configuration Requirements for DMP Processing
4.1.1. Prerequisites for Running a DMP Analysis
4.1.2. Setting Up the Cluster Environment for DMP
4.2. Starting a DMP Analysis
4.2.1. Starting a DMP Analysis via the Launcher
4.2.2. Starting a DMP Analysis via Command Line
4.2.3. Starting a DMP Analysis via the HPC Job Manager
4.2.4. Starting a DMP Analysis via a Linux Job Scheduler
4.2.5. Starting a DMP Solution in the Mechanical application (via Ansys Workbench)
4.2.6. Using MPI Files
4.2.7. Directory Structure Across Machines
4.3. Supported Analysis Types and Features
4.3.1. Supported Analysis Types
4.3.2. Supported Features
4.4. Understanding the Working Principles and Behavior of Distributed-memory Parallelism
4.4.1. Differences in General Behavior
4.4.2. Differences in Solution Processing
4.4.3. Differences in Postprocessing
4.4.4. Restarts with DMP processing
4.5. Example Problem
4.5.1. Example: Running a DMP Analysis on Linux
4.6. Troubleshooting
4.6.1. Setup and Launch Issues
4.6.2. Stability Issues
4.6.3. Solution and Performance Issues
5. Using Hybrid Parallel Processing (SMP + DMP)
5.1. Activating Hybrid Parallel Processing
5.1.1. Activating Hybrid Parallel by Command or Product Launcher
5.1.2. System-Specific Considerations