This article describes the workflow and settings/considerations for running ANSYS Platform on NIMBIX – computationally intensive processes (large models).
For transferring CAD files (for large assembly simulations), use FileZilla (free download from the web). Use your NIMBIX username and your API key from the web portal to access your account. If you need a refresher on FileZilla, click here.
FileZilla can also be used to retrieve archived files from the cloud or entire folders that contain model and results data files. Simply drag and drop files from one location to the other (connecting to Jarvice, /home/nimbix/data folder allows you to access the model and solution files from the cloud)
To access ANSYS Platform on NIMBIX platform, the following steps are required:
- Select the desired ANSYS Platform release from the Compute dashboard
NOTE: If the option is not available in the first page menu, press on “More” at the bottom of the page as shown in the image below:
2. A splash window will open. Select the ANSYS Workbench Platform option as shown below:
3. Cloud set-up screen opens and here you must choose some of your settings by clicking on the Tabs on the top of the window (General, Optional, etc) one tab at a time.
UNDER GENERAL TAB
1. Under Machine type when you click on the drop down menu, you can select the type of machine you want to run your job on. The decision on machine type selection is based on size and complexity of your model and cost associated with the machine type (some machines will have higher RAM, others will only run the job on single CPU, others will have better graphics and therefore higher cost, etc).
NOTE: When running interactive based applications, you’ll find that selecting an NC9 or any NC* machine types should offer significant visual performance over not selecting an NC machine type. By selecting an NC machine, this places a GPU on your head-node, and offers better visual performance. Another thing to keep in mind, is that when running interactively you can use a web-browser, or in some cases for large models or you might consider using RealVNC.
2. Select the number of cores:
The machine type you selected in the previous step, will dictate the increment in the number of cores that you can choose/select. For a very simple and small model, you can leave default selection, which in this case would be “16” or move the scroll bar to the desired number of cores or simply type over “16” the number of cores you wish to run your job on (we left it default in this case):
NOTE: Do not confuse number of cores with number of nodes (nodes represent the number of increment of cores that you selected. In the example above, 1 node represents 16 cores, 2 nodes correspond to 32 cores).
UNDER OPTIONAL TAB
1. Assign a JOB LABEL (give a name that will help you keep track on your running jobs. For example, MyPlatformProject):
Leave blank the wall time limit and the IP address. The Window size needs to be kept as default.
UNDER STORAGE TAB
1. Select vault type: Default vault is “Elastic_File
The “Elastic_File” vault is recommended for small to medium size jobs, such as Icepak projects, simple linear Mechanical Analysis projects, some HFSS and simple Fluent projects (not multi-phase). For any complex and computationally heavy jobs, and where partitioning the job over number of cores becomes challenging, the Performance_SSD vault is strongly recommended. The Performance_SSD vault can be found in the drop-down under “Select Vault” tab (NOTE: requires subscription and extra monthly payment to have access to Performance_SSD vault).
Before submitting your job for running, you can preview your settings under the PREVIEW SUBMISSION tab.
2. You can start your job by clicking on the SUBMIT tab.
After ANSYS Workbench Platform has been successfully launched, you should be able to see the ANSYS Workbench in the preview and can access the window by clicking on it (you can see details on the number of nodes, some information regarding memory, etc):
After clicking “OK” on ANSYS feedback survey participation window, you are inside the Ansys Workbench environment:
ANSYS Workbench is a software environment for performing structural, thermal, and electromagnetic analyses. You can also perform tasks such as geometry creation and optimization, attaching an imported CAD geometry, setting up the finite element model, solving, and reviewing results.
Be sure to save this project by clicking on “File” “Save As” MyPlatformEx1 (any name you wish to give; the name here is your file name, not be confused with the job name which is just a bookkeeping name of the project that runs, especially when you run multiple projects at once).
BUILD YOUR FINITE ELEMET MODEL: Most large models contain several linked analyses and other system components such as Design Exploration. In this example a Modal analysis is linked to a Structural analysis and a response surface DOE and an optimization section (see figure below). Mechanical analysis can be set-up using “Drag-and-Drop” methodology in ANSYS Workbench and the parametric study follows the parameters workflow in ANSYS Mechanical/Workbench.
NOTE: The number of cores selected at the beginning of the process is shown in the “Solve” toolbar. Do not increase the number of cores once you selected them in the start-up process.
NOTE: Additional analysis environments can be also added to the active/current ANSYS Mechanical analysis through the “Analysis” option (can be used for Multiphysics platform analysis or pre-load analysis). See figure below:
RUN YOUR ANALYSIS: Monitor the progress in the “Detailed Job Metric” window by clicking the corresponding icon in the Dashboard for the job selected.
NOTE: Some of the processes (meshing, modal analysis) may not use 100% of your allocated CPU (see image below) as shown.
As the process becomes more intensive, the CPU utilization and memory usage are updated (see image below). If the running times are too long one can add additional cores for subsequent runs.
VISUALIZE YOUR RESULTS: Save your run and use a new session (less computationally intensive) to visualize the results. Use NC* machines to get the best graphical representation with less cores (unless you intend to make changes to your model and re-run):