Distributed Processing

Description

Distributed processing supports multi-machine distributed operations for handling large volumes of point cloud data files.

Usage

Assuming you have four computers: one acting as the master node and the other three as computing nodes.

1.Set Up and Start Computing Nodes

On the three computers designated as computing nodes, start lislave.exe. In the pop-up settings dialog, enter the IP address and port number for each computer. The port number can generally be left at its default value. After clicking the OK button, a computing node monitoring dialog will appear on each of the computing nodes.

2.Set Up and Start Master Node

Once the three computing nodes are running, click the "Distributed Computing" button on the Lidar360 toolbar to open the master node settings dialog. Use this dialog to configure distributed tasks, which involves three main aspects:

(1) Configure IP and Port for Each Computing Node

Ensure that the IP addresses and port numbers configured here match those entered in the computing node settings dialog mentioned earlier.

(2) Configure Computational Task

For example, you can configure the tasks to "Classify ground points first, then perform resampling."

(3) Set Source and Result Point Cloud Directories

The source point cloud directory contains the point cloud data to be processed, which can include multiple large data files. The results of the distributed processing will be output to the result point cloud directory.

Start Distributed Computing Process

Click the Execute button to open the master node monitoring dialog and start the distributed computing process.

Monitoring Distributed Tasks

In the master node monitoring interface, the top list displays real-time information about each source point cloud file chunk, including the IP of the computing node it's assigned to, the current stage of the chunk, the status of the chunk, and the time consumed for processing the chunk. The middle list shows real-time information for each computing node, including connection status, point cloud transfer speed, CPU usage, memory usage, the number of current Libatch instances running on the node, and the maximum number of Libatch instances run historically. The lower section displays log output and overall progress information for the distributed computation.