Hardware Benchmarking (2018-03-AA)

Introduction

We frequently get asked, "What is the minimum or recommended hardware to use for TUFLOW modelling". This is always a tricky question, as the answer depends on the type and size of the models you are going to run. For a small model, TUFLOW should run on any modern PC or laptop that is capable of running Windows XP or later. However, for large models there may be requirements for a hefty computer.

The tables below showing computer specifications and model run-time should help you compare systems.

In this page we outline a hardware benchmark model which is available to download from FMA2_GPU_CPU_Benchmark.zip the model can be simulated without a TUFLOW dongle (license). This makes it easy to benchmark on a range of computers and the results are compiled below.

We have typically found that for TUFLOW Classic, the CPU speed is the largest influence on TUFLOW runtimes, with the RAM speed also having an influence for large models. When using GPU hardware (available for TUFLOW HPC), the Graphics Card has the largest influence, specifically the number and speed of the CUDA cores. In order to quantify this we are compiling the computational times required for a range of different machines.

Benchmark Model

Background

The benchmark model is based on a “challenge” issued prior to the 2012 Flood Managers Association (FMA) Conference in Sacramento, USA. There is more information on the model setup and purpose in the FMA challenge model introduction.
This hardware benchmark is based on the second challenge which involves a coastal river in flood with two ocean outlets. The model has been modified slightly (mainly in terms of the outputs). It is setup to use both the TUFLOW Classic (CPU) and TUFLOW HPC (on both CPU and GPU hardware) for a range of cell sizes.

Cell Size (m)	Number of cells
20.0	181,981
10.0	727,865
5.0	2,911,472
2.5	11,645,341

The model runs for three days of simulation time (72 hours). The approximate run time for the 30m model on the CPU is likely to be ~20min and for the 15m version approximately 4 hours. Given the runtime for the CPU model at 10m resolution is likely to be > 12 hours, this is skipped in the benchmark (this can also be run without a licence).

Models

The simulated models are:

• 20m model - TUFLOW Classic
• 20m model - TUFLOW HPC CPU - 2 Cores (default)
• 20m model - TUFLOW HPC - GPU Hardware
• 10m model - TUFLOW HPC - GPU Hardware

The user also has an option to run a couple of bigger models on GPU hardware as part of the High End GPU Benchmarking. To run these models, please follow the steps outlined in the below section.
HIGH END GPU BENCHMARKING MODELS

• 5m model - TUFLOW HPC - GPU Hardware
• 2.5m model - TUFLOW HPC - GPU Hardware

To Participate

To participate in the benchmark, please follow the steps below:

1. Download the model from https://www.tuflow.com/Download/TUFLOW/Benchmark_Models/FMA2_GPU_CPU_Benchmark.zip
2. Extract the model on a local drive of the computer you would like to benchmark.
3. Navigate to the TUFLOW\runs\ folder and run the Run_Benchmark.bat file. This checks if you are running a 32 or 64 bit system and then runs the benchmark. This also generates some output files that contain more information on the processor, memory and GPU card that you are using.
4. If you have a high performance GPU Card you can also run the Run_Benchmark_HighPerformance_GPU.bat.
5. Email the _ TUFLOW Simulations.log, cpu.txt, ram.txt and GPU.txt files to support@tuflow.com and we will includes these in the results tables below.
   

The computer information is determined in the batch file using the wmic and dxdiag commands.

Note An NVIDIA graphics card that is CUDA compatible is required to run the GPU model.

List of Benchmarking Runtimes and Discussions

Benchmarking Runtimes (2018-03-AA Release)

• CPU results
• GPU results

Discussion Topics

• Classic vs. HPC (Coming Soon) (notes: direct rainfall, dam break, runtimes)
• HPC (CPU) vs. HPC (GPU)
• HPC (CPU) Scaling and Hyperthreading (Coming soon)
• HPC (GPU) Scaling (Coming soon)
• Single Precision Vs Double Precision
• Influence of 1D on Runtimes (Coming soon)
• Hardware Selection Advice
• Spectre and Meltdown

Previous Benchmarking Runtimes (2013-12-AD)

• Previous Computer Hardware Speed Benchmarks Archive
• Previous CPU Results
• Previous GPU solver results