Notes on Different Formats

If your data format is not currently supported by Vapor you may be able to convert your data from its native file format to one that is supported. Several of Vapor’s supported formats may be good candidates for your conversion. The table below shows these formats, along with their supported and unsupported features.

	CF Compliant netCDF	BOV	VDC	DCP
Regular grids	Yes	Yes	Yes	No
Rectilinear grids	Yes	No	Yes	No
Curvilinear grids	Yes	No	Yes	No
Missing values	Yes	No	Yes	No
Raw, Binary data	No	Yes	Yes	No
Multi-resolution*	No	No	Yes	No
Complexity	Medium	Low	High	Medium
Particle Data	No	No	No	Yes

For example, if your data is organized as a collection of RAW files that are sampled on a regular grid you may want to use Vapor’s BOV format. This is the easiest way to import unsupported data because it only involves creating a small, ASCII metadata file that describes your data.

Note

The geophysical modelling community often uses the term “stretched grid” in place of “rectilinear grid”. In Vapor parlance, they are the same.

If your data are sampled on rectilinear (a.k.a. “stretched”) or curvilinear grids, or if your grid is regular and stored in a format other than raw-binaries, converting your data to CF compliant NetCDF is probably your best bet. CF NetCDF is widely used in the earth sciences and is highly flexible. There is a large ecosystem of software tools that are capable reading, writing, and manipulating CF NetCDF files.

If your data are sampled on a high resolution grid, and performance is a concern, you may want to consider VAPOR’s multi-resolultion VDC file format. However, this format should only be considered if multi-resolution is required. For small to modest sized data sets, performance may be degraded by using the VDC.

Note

The multi-resolution feature of the VDC Allows you to render large datasets with or without lossy compression. Viewing compressed data can dramatically increase the interactivity for your visualizations if your computer’s resources are constrained by the size of your data.

When you’re done with exploring compressed data and ready for a final rendering, you can render an image or animation sequence with lossless compression.