Hardware Components Usage when processing with Pix4D

 
Information: 
  • For more information about the use of the GPU: 203405619.
  • For recommendations for a Hardware and Software Configuration: 202559159.
  • For more information regarding: Mac / Windows XP / Linux / Remote Access - Virtual machine / Distributed - Parallel processing: 202556809.
  • For more information about the processing speed: 204191535.

 

The 4 key components of the computer are:

  • CPU
  • RAM
  • Hard Disk
  • GPU

The software is highly parallelized and takes advantages of multi-core CPUs, as well as MMX/SIMD instructions and NVIDIA GPU Cuda processing.

However the different steps of the processing do not use the resources the same way, and not all parts can take advantages of multi-core or Cuda.

To improve the performance it is important to have a balanced configuration in order not to have bottlenecks.

 

CPU

Most of the processing is done in the CPU. A faster CPU is the first key to increasing processing speed.

Hexa and octa cores, latest generation i7 or Xeon CPUs are recommended. Clock speed generally impacts the full project, and the number of cores impacts step 2.

Dual socket CPUs do not double the performance, but they do generally provide faster processing than single socket CPUs.

 

RAM

The amount of RAM mostly has an impact on the number of images that can be processed in a single project, and less on the processing of step 2.

 

Hard disk

Step 3 is the most influenced by the speed of the hard disk. When using octa-core CPUs and a fast SSD the performance of step 3 increases.

 

Graphic Card

Starting with version 1.3, Pix4Dmapper takes advantages of NVIDIA GPUs with CUDA to further increase processing speed.

The speed increase highly depends on the project type including image number, image size and image content. For example, with a GTX 970 a speedup between 10% and 75% on step 1. Initial Processing, and around 10% on step 2. Point Cloud and Mesh was observed. As a rule of thumb, projects with high overlap, high image content and thus a high number of keypoints benefit more from the speedup.

 

Step 1      
CPU RAM Hard Disk GPU
Most important component.

In some sub-processing steps all the available CPU resources will be used whereas in others not all the CPU is used*.

Low usage. Low usage. (optional) Medium/high usage.
Step 2      
CPU RAM Hard Disk GPU
Most important component.

Fully used. This step will use all the CPU resources available.

Most important component.

Fully used. This step will use all the RAM resources available.

Low usage. Low Usage 
Step 3      
CPU RAM Hard Disk GPU
Low usage. High usage. Most important component.

The speed of the Hard Disk defines the processing speed.

No usage.
rayCloud      
CPU RAM Hard Disk GPU
Low usage. Low usage. Low usage. Most important component.

 

*Some parts of the processing cannot be paralleled and some of them do not consume a lot of resources. However, they need to be completed in order to start the next processing process.

Was this article helpful?
5 out of 5 found this helpful


Powered by Zendesk