Designing an efficient cloud particle system to represent weather

My goal is to make something that looks like the Gibson Ridge level 2 analyst: GR2Analyst Main Page

103423-grl2.png
I would like to create a system that represents a cloud, but not in the usual video game sense. The general idea is that I get data from the National Weather Company. It comes a format where each line of a text file is the following:
x,y,z,data (for example 100.1, 500.3, 23.2, 50.0)

The x,y,z are data points relative to a radar. The “data” component is the wind velocity returned by the radar at that point. The program will cut off anything farther than 100km from the radar.

The radar will return about 8 million data points (lines of x,y,z,data) every minute, returned in a text file. So, efficiency is of the utmost importance. I tried to do the easiest way of spawning cubes at each location with the color based on the data value. Obviously that isn’t the best way, so I’ve been looking into particle systems. Basically I need to create some form of mesh(not sure how this would work with colorizing correctly) or particle system that represent the data. Particle systems have proven difficult as I need to create 8 million static(the particles don’t need to move, I just need to be able to place the particle system at some arbitrary point and have it spawn the cloud) particles, and set each one’s position/color.

Can anyone offer a solution that I could look into that can help me create this large system? I can supply more info if necessary. I would think that turning this into 1 mesh, or using a particle system would be the best way, but maybe I’m missing something.

Look into Point Meshes. They’re used for voxel-based rendering and would be ideal for this application.