Hardware Selection Advice: Difference between revisions

We often get asked about the optimum computing setup to run TUFLOW models. While every model is different and will interact differently with your hardware there is some general advice that we can offer. InNote thethat sectionsrecommendations belowfocus youspecifically willon findrunning moreTUFLOW. detailedIt adviceis onhighly GPUrecommend to consult your IT team to confirm that all components of the machine are fully capable of supporting your intended uses and CPUmeet butyour requirements for quality, speed, and generally:durability.<br>

* The system must be well cooled to avoid throttling (meaning reduction of clock speeds to reduce heating), and have sufficient and reliable power supply. Should upgrades to the system be expected in the future (such as adding a second GPU card), then consider configuring these components to avoid future limitations. <br>

For information on minimum and recommended system requirement, see <u>[[System_Requirements | System Requirements]]</u>.<br>

* TUFLOW Classic: A single model run can only use the CPU and cannot be run across multiple CPU cores or GPU hardware. In general terms: The maximum model size is dependent on the available CPU RAM and the runtime is driven by the CPU speed, architecture and cache size.

* TUFLOW HPC - Run on GPU Hardware: A single model run uses the GPU(s) for computation. In general terms: The maximum model size is dependent on the available GPU and CPU RAM and the runtime is driven by the CUDA core speed, the number of CUDA cores available and the GPU architecture. GPU performance is complex and is not easily inferred from GPU clock speed and number of cores, it is also very dependent on the ‘generation’ or architecture of GPU. As TUFLOW HPC requires some data exchange between GPU and CPU, the motherboard bus speeds and CPU speeds also play a role but typically a much lesser role compared to the GPU CUDA compute.

* TUFLOW FV - Run on CPU Hardware: A single model run uses CPU and is parallelised to run across multiple cores. In general terms: The maximum model size is dependent on the available CPU RAM and the runtime is determined by the CPU speed, the number of cores available to be run in parallel, chip architecture and cache size.

* TUFLOW FV - Run on GPU Hardware: A single model run uses the GPU(s) for computation. In general terms: The maximum model size is dependent on the available GPU and CPU RAM and the runtime is driven by the CUDA core speed, the number of CUDA cores available and the GPU architecture. GPU performance is complex and is not easily inferred from GPU clock speed and number of cores, it is also very dependent on the ‘generation’ or architecture of GPU. As TUFLOW FV requires some data exchange between GPU and CPU, the motherboard bus speeds and CPU speeds also play a role but typically a much lesser role compared to the GPU CUDA compute.<br>

On ourThe <u>[[Hardware_Benchmarking_-_Results#CPU_Results | Hardware Benchmarking]]</u> page you can compareshows recently run combinations of GPU, CPU and RAM. These can be compared with the system youintended are planning tofor purchase. WeThe recommendrecommendation thatis if building a computer that youto seek advise from an appropriate computer hardware vendor who can advise on the compatibility and optimisation of yourthe setup.<br>

* TUFLOW HPC on GPU Hardware can be run in either single or double precision. However, for the vast majority of flood applications single precision is sufficient. We typically run our models on single precision. If you are unsure we recommend running with both the single and double precision solvers and comparing your results.

TUFLOW HPC on GPU hardware still uses the CPU to compute and store data (in CPU RAM) during model initialisation and for all 1D calculations. While we are working on improving our CPU RAM usage, currently we tend to find that CPU RAM is often the limiter to the size of the model domain you can run, particularly if using running over multiple GPU cards. During initialisation and simulation a model will typically require 4-6 times the amount of CPU RAM relative to GPU RAM. As an example, a model that utilises 11GB of GPU RAM (typical memory for high-end gaming card, and corresponds to about a 50 million cell model) the CPU RAM required during initialisation will typically be in range 44GB to 66GB. A model that fully utilises two 11 GB GPUs (i.e. a 100 million cell model) may require as much as 128GB of CPU RAM during initialisation. Note that anything more than 256GB of CPU RAM will exceed the limitations of consumer chipsets that are available in 2025 and requires more expensive workstation hardware - additionally, users should consult a hardware expert to check limitations of specific hardware.

===CUDA Cores, GPU Clock speed, and FLOPs ===

* TUFLOW HPC-GPU does not support SLI for inter-GPU communications.

* It does (as of build 2020-01-AA) auto detect and utilise peer-to-peer access over NVLink or PCI bus on the motherboard. Note that not all GPUs support peer-to-peer access.

** PCI bus - this method requires cards that supports TCC driver mode and all cards must be in TCC driver mode. As TUFLOW primarily relies on GPU CUDA capabilities, the impact of using higher or lower PCI slot option is minimal.

** NVLink - high-end compute cards can have up to 8 cards talking to each other through a high-spec NVLink, but many of the less expensive cards are limited to only having two connected together over a dual socket NVLink.

* Models may still be run across multiple GPUs even if a NVLink is not present and the GPUs do not support peer-to-peer access. In this case HPC reverts to exchanging the domain boundary data between the GPUs via the CPU. The memory bandwidth between the GPU and the main system is not a critical bottleneck for TUFLOW.

* When using multiple GPUs it is best to use cards of similar memory and performance. While it is possible (as of build 2020-01-AA) to re-balance a model over multiple GPUs, we do not recommend using cards with vastly disparate performance.

* Sufficient cooling and power supply should be considered if multiple cards are used. When installed in adjacent PCI slots, the preference is to use rear vented cards rather than side vented to avoid blowing hot air onto the neighbouring cards (which could lead to overheating).

Extensive GPU hardware speed comparison testing has been completed using TUFLOW's standardised hardware benchmarking dataset. Details for the benchmarking are available via the <u>[[Hardware_Benchmarking_(2018-03-AA)Hardware_Benchmarking | Hardware Benchmarking]]</u> page of the Wiki. Review the GPU benchmarking runtime results table to compare the speed performance of different cards. If your GPU card is not listed in the result dataset please download and run the benchmarking dataset, and provide the result summary to [mailto:support@tuflow.com support@tuflow.com]. We will add the details to the runtime results table.<br>

External videocard benchmark websites can be used to compare GPU cards, for example, <u>[https://www.videocardbenchmark.net/high_end_gpus.html PassMark Software - Video Card (GPU) Benchmarks]</u> is an excellent performance guide. Note that PassMark may not be representative for the highest end cards, for TUFLOW. GPU are complex devices, newer cards may not perform as well on PassMark's benchmarks for criteria consumers buy GPUs for (games, video editing, etc.), even though the cards may well perform a lot better for TUFLOW.

===CPU Cores ===

* TUFLOW HPC - Run on GPU Hardware: The parallel processing is being done on the GPU card. However, TUFLOW HPC-GPU still uses the CPU for model initialisation and for 1D calculations. If multiple GPU cards are used, TUFLOW will use the equivalent number of CPU threads for controlling the GPUs and migrating data. So for a machine dedicated to HPC-GPU modelling, the number of CPU cores should be higher than the number of installed GPUs.

* TUFLOW HPC - Run on CPU Hardware: HPC model can also be run on multiple CPU cores. For the comparison of simulation speed, please refer to [[Hardware_Benchmarking_Topic_HPC_on_CPU_vs_GPU | HPC on CPU vs GPU]].

* TUFLOW Classic: TUFLOW Classic simulation can only use one CPU core due to the implicit nature of the numerical solution. More CPU cores will enable running more simulations at the same time most efficiently.

Extensive CPU hardware speed comparison testing has been completed using TUFLOW's standardised hardware benchmarking dataset. Details for the benchmarking are available via the <u>[[Hardware_Benchmarking_(2018-03-AA)Hardware_Benchmarking| Hardware Benchmarking]]</u> page of the Wiki. Review the CPU benchmarking runtime results table to compare the speed performance of different chips. If your chip is not listed in the result dataset please download and run the benchmarking dataset, and provide the result summary to [mailto:support@tuflow.com support@tuflow.com]. We will add the details to the runtime results table.<br>

Solid state hard drives are preferred for temporary storage as they are faster to write to than traditional hard drives. Large data files can then be transferred to a more permanent location.<br>