GP2PCF and GP2SSCF

Fast 2 point correlation functions using GPUs

Home

 

DOWNLOAD

 

LICENSE AND oldER vERSIONS

 

About Us and references

 

Contact Us

 

 

Fast 2 point correlation functions using GPUs

 

GP2PCF: a code for brute-force computation of 2 point angular correlation functions

GP2SSCF: a code for brute-force computation of shear-shear correlation functions

 
 
 

Recent Updates

 
December, 2012 GP2SSCF: The first shear-shear correlation function public version (V 0.2)

a code for brute-force computation of shear-shear correlation functios

 
October, 2012 Version 0.4 of GP2PCF code
Include improvements such as registers utilization into the kernel implementation
 
February, 2012 GP2PCF: The first 2 point angular correlation function public version (V 0.2)
a code for brute-force computation of 2 point angular correlation and cross-correlation functions
 
 
 

Download code

 

GP2PCF Code

 

Current CUDA and PyCuda Version 0.4 features:

1) 2 point angular correlation function computation implemented in CUDA (auto and cross correlation)

2) The Code was updated using registers in its kernel in order to be faster.


GP2PCF CUDA version 0.4

GP2PCF PyCuda version 0.4

 

IMPORTANT: Sometimes you can note a difference between our code to others at small angles, it could be due to two reasons:

1) k-trees or pixelitation can not reach small scales with enough accuaracy and

2) the arc-cosine calculation is not very precise for numbers really close to one.


 

GP2SSCF Code

 

Current CUDA version 0.2 features:

1) Shear-Shear correlation function computation implemented in CUDA

 

GP2SSCF CUDA version 0.2

GP2SSCF PyCuda version 0.2 (In Progress)

 

IMPORTANT: The results has been compared with athena code using its input data set, in order to compare please look at the README file.

 

 
 

 

 
2013. Designed by Web Page Templates
 
 
Design downloaded from free website templates.