Skip to content

mfkiwl/FLAG-Beamformer-Devel

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

FLAG-Beamformer-Devel
This is a fork of the Vegas development repository. It will contain Vegas code that has been modified to support the development of the FLAG-Beamformer project.

This repository is intended to be used alongside the FLAG-Beamformer-Simulator-I repository. 

A brief overview of the contents as ther pertain to the FLAG-Beamformer project:

We are using this repo to accomplish two main things: NRAO-specific FITS formatting and easy usage of the Dealer/Player implementation provided by DIBAS. This has allowed us to very easily make a proof-of-concept Beamformer FITS implementation, although the code requires siginificant work to reach production.

The NRAO's ygor library contains a FitsIO class that wraps the cfitsio library and outputs FITS files that conform to NRAO standards. This repo contains a FITS writer that subclasses FitsIO and adds Vegas-specific functionality to it. This FITS writer has been refactored and modified to output FLAG-Beamformer-specific FITS files.

The VegasFitsIO subclass defines the FITS operations, while the VegasFitsThread is responsible for running the scan and making the write calls.

Most of the change has ocurred in dibas_fits_writer, notably:
	main.cc
	VegasFitsIO.{h,cc}
	VegasFitsThread.{h,cc}

There were also changes to some of the vegas_hpc code, notably:
	vegas_databuf.{h,c}

The vegas_databuf struct is a 1:1 copy of the gpu_output_databuf struct over in the hashpipe repository. Eventually it would obviously make more sense to inherit the type(s) from a common header, but this led to a quicker proof-of-concept.

Note that vegas_hpc contains C code that is very, very similar to hashpipe code.
Akshaya was here.

About

This is a fork of the Vegas development repository. It will contain Vegas code that has been modified to support the development of the FLAG-Beamformer project.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 52.7%
  • C 28.2%
  • C++ 11.2%
  • TeX 3.3%
  • Cuda 2.0%
  • Shell 1.8%
  • Other 0.8%