
Scale-Resolving Simulation at Scale
A complete guide to running GPU-accelerated simulations on our platform
SRS is a cloud-based simulation platform that leverages the high-order PyFR engine, GPU acceleration, and AI to deliver faster, more efficient Scale-Resolving Simulations of turbulent industrial flows. Before you begin, here's what you need to know:
A valid email address for account creation
Basic understanding of simulation parameters, our AI will guide you!
A valid credit card
Setting up your SRS account is quick and straightforward. Follow these steps to get started:
Create Your Account
Click "Get Started" on the landing page and fill in your details. You'll receive 2 hours of free compute time which you can buy just with 1€ + VAT!
Verify Your Email
Check your inbox for a welcome email.
Sign In
Use your email and password to access your dashboard.
Explore Your Dashboard
Familiarize yourself with the interface — your compute time, storage, and quick actions are all accessible here.


Now for the exciting part — launching your first GPU-accelerated simulation!
Navigate to Start Simulation
From your dashboard, click the "Start Simulation" button or use the sidebar menu.
Describe Your Simulation
Give your simulation a name and brief descriptive so you can easily identify it later.
Prepare your ZIP folder
Our platform takes a zip folder containing the simulation configuration file which must be exactly named "pyfr.ini", plus any mesh files your case requires (e.g. a Gmsh .msh mesh).
Validate your pyfr.ini file
Inorder reduce the risk of errors, you must validate your pyfr.ini file before uploading your ZIP folder. Click the "Validate pyfr.ini file" button to validate your pyfr.ini file. For free users, the validation only checks the basic syntax. For paid users, the validation process is more comprehensive and includes the syntax check, system check, mesh check, and AI analysis for simulation parmeter stability. See section on AI features for more details.
Upload Your ZIP folder
Click "Choose File" to select your ZIP folder. The folder will be uploaded to your cloud storage.
Review and Submit
Double-check your settings and click "Start Simulation". Your simulation will be queued and start running on our NVIDIA A100 GPUs.



Track your simulation progress in real-time through the Jobs Portal.
Your job is waiting in line
Performin system check, grid check, assigning resources and setting up the environment.
Simulation is actively running on GPU
Simulation finished successfully
An error occurred (check logs for details)
Simulation was cancelled by the user
Estimated Time Remaining (ETA)
Predicted completion time based on current progress
Duration
Total time which includes all the stages from enqueue to completion and processing time which excludes the enqueue time. Hours consumped as calculed from processing time only.
File Size
Size of the resulting files in ZIP format
Action Buttons
View, download, cancel, or delete the job. Logs are available for viewing.
Total Consumption
Total and reaming consumed hours are shown via the progress bar.



Once your simulation completes, you can download the results for analysis.
Go to Jobs Portal
Navigate to the Jobs section from your dashboard.
Find Your Completed Job
Look for jobs with the "COMPLETED" status badge.
View Job Details
Click on the job to expand details and see output files.
Download Results
Click the "Download" button to save result files to your computer. You can also view logs directly in the browser.
Solver output files (PyFR .pyfrm data + VTK/PVTU exports)
SRS solver logs with detailed runtime information
Performance metrics and timing data
The PyFR engine writes results in its native .pyfrm format and can export to VTK/PVTU for use with third‑party visualization tools such as VisIt (developed by Lawrence Livermore National Laboratory, LLNL) and ParaView (developed by Kitware, Inc.) to inspect and analyze your SRS output. Configure the [output] section of your pyfr.ini with format = pyfr, vtk to write both formats simultaneously.
Viewing SRS simulation results in VisIt involves opening the VTK/PVTU files exported by PyFR, then using VisIt's plotting and analysis tools. Below is a concise guide to get you started.
Install a compatible VisIt version for your operating system (VisIt supports multiple platforms and data formats).
Confirm that your SRS output is exported in a VisIt-readable format. PyFR's VTK/PVTU export is the most reliable option; XDMF is also supported. If unsure, check the PyFR documentation for export options.
Prepare your data path or database: point VisIt to the directory containing the VTK/PVTU files. VisIt can also aggregate multiple time steps if you open the group as a database.
Start VisIt and open a database
Go to File → Open File (or Open Database) and select your PyFR-exported VTK/PVTU file or directory.
Create a plot
Choose an appropriate plot type (e.g., Pseudocolor for scalar fields like pressure or vorticity magnitude, Vector for velocities).
Apply data operations
Use Expressions to compute derived quantities (e.g., Q-criterion, λ2, or custom scalars for turbulence visualization).
Time and animation
If your results contain time steps, use the Animation controls to explore temporal evolution of turbulent structures.
Save visuals
Export images or animations via File → Save Window or Save Animation.
ParaView is another widely used post-processor for PyFR-exported VTK/PVTU outputs. The workflow is similar to VisIt but uses its own UI and readers.
Install a compatible ParaView version
Download a version appropriate for your OS. Ensure that the release includes the VTK readers you need for HPC data formats.
Confirm the SRS output format
PyFR's VTK/PVTU export is the most ParaView-friendly option. Set format = vtk in the [output] section of your pyfr.ini if you only need ParaView compatibility.
Load the data in ParaView
Go to File → Open, then select the PyFR-exported VTK/PVTU file or directory. ParaView will load the dataset and display available fields. For multi-step outputs, open the group as a time-varying dataset rather than individual time-step files.
Visualizing fields
Use the Properties panel to choose a representation (e.g., Volume, Surface, Glyphs). Adjust color maps and opacity to highlight important variables (pressure, velocity magnitude, vorticity).
Time steps and animation
If the dataset includes multiple time steps, use ParaView's animation controls to play or scrub through time. ParaView automatically exposes time information for temporal datasets.
Derived quantities
Use the Calculator or Expressions to compute derived fields (Q-criterion, λ2, vorticity, etc.), then visualize them alongside primary fields.
Saving and exporting results
Use File → Save Screenshot or Save Animation to export images or movies. You can also save a ParaView state file to reproduce your visualization setup.
format = vtk in the [output] section of your pyfr.ini to produce ParaView-readable files. Review the PyFR documentation for recommended export formats.Get the most out of SRS with these expert tips:
Ensure your simulation parameters are well-defined to avoid wasted compute time. Start with smaller test cases before scaling up.
Name your jobs clearly (e.g., "shock_tube_high_res_v3") so you can easily track multiple simulations.
Check your jobs periodically to catch any issues early. Premium users get AI-assisted error detection.
While storage is unlimited, keep your workspace organized by deleting old jobs you no longer need.
SRS goes beyond traditional simulation platforms by integrating advanced AI capabilities to enhance accuracy, catch errors early, and optimize your simulations. Available exclusively for Premium users.
Our AI performs comprehensive validation of your pyfr.ini file beyond basic syntax checking:
Syntax Check: Verifies Python code structure
System Check: Validates simulation parameters and configurations
Grid Check: Analyzes mesh quality and resolution
AI Stability Analysis: Predicts potential numerical instabilities
Get AI-powered suggestions to optimize your simulations:
Parameter Optimization: Suggests optimal time steps, convergence criteria
Resource Allocation: Recommends compute resources based on problem size
Stability Warnings: Flags potentially unstable configurations
AI validates your simulation setup against physical principles:
CFL Condition Check: Ensures time step meets stability criteria
Boundary Consistency: Validates boundary condition compatibility
Material Property Check: Flags unrealistic physical parameters
When simulations fail, AI helps you understand why and how to fix it:
Error Analysis: Interprets error messages in plain language
Root Cause Detection: Identifies underlying issues
Solution Suggestions: Provides actionable fixes
Upgrade to Premium
Purchase any credit plan (Standard, Professional, or Enterprise) to unlock AI features.
Validate with AI
When starting a simulation, click "Validate pyfr.ini file" to trigger comprehensive AI analysis.
Review AI Insights
Check the validation report for warnings, recommendations, and stability predictions.
Monitor Runtime Errors
AI continuously monitors your jobs and provides a comprehensive diagnosis of the error. It provides the error type, root cause, and the corrected code itself.


Catch errors before running simulations, saving hours of wasted compute time
AI optimization ensures more stable and accurate simulation results
AI explanations help you understand and improve your simulation skills
Now that you know the basics, it's time to run your first simulation and experience the power of GPU-accelerated computing with AI assistance.
© 2026 SRS, a NumericalAI product by Empirisch Tech GmbH (empirischtech.at). All rights reserved. |Privacy Policy |Terms of Service |Executive brief |FAQ