Skip to content

kmsmith137/n2k

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

143 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

CUDA kernels for chord (tensor-core visibility matrix, fast rfi flagging)

The n2k library contains:

  • An int4+4 GPU visibility matrix kernel with a boolean RFI mask.

  • GPU kernels which implement the high-speed RFI masking logic in CHIME/CHORD.

Some loose ends that I might fix later. Let me know if you'd like me to prioritize any of these. See n2k.hpp for more info:

  • Currently, we only support 4+4 bit electric field samples, in the range [-7,7]. If the value (-8) arises, then the output of the computation will be incorrect!

  • The kernel will segfault if run on a GPU which is not the cuda default device.

The n2k library currently doesn't use a proper build system. It assumes that you've built the ksgpu library (https://github.com/kmsmith137/ksgpu) in a sibling directory ../ksgpu. Then do the following:

git clone https://github.com/kmsmith137/n2k
cd n2k
make -j
./bin/test-correlator

I'll improve this this in the future!

Contact: Kendrick Smith [email protected]

About

CUDA kernels for chord (tensor-core visibility matrix, fast rfi flagging)

Resources

Stars

3 stars

Watchers

3 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors