Page History

...

• test common-mode speed on gpu (seshu has demonstrated Mikhail’s calibration formula without common-mode will be fast on a GPU)
• we should meet with XPP scientists to understand what data-reduction algorithms are needed for their hutch
• if libsz is one of the algorithms, we should understand its performance on gpu
• benchmark other data-reduction algorithms on gpu
• consider multiple options for algorithm implementation: cupy, cunumeric, cuda kernels
• talk to TID engineers about dma'ing the kcu1500 (or c1100) data directly to gpu

Fred wrote: Regarding the bullet point “we should meet with XPP scientists to understand what data-reduction algorithms are needed for their hutch”, I just wanted to share an initial list since Valerio and I met with Vincent recently and what stood out was:

• ROI (available in smalldata_tools)
• Photonization (available in smalldata_tools; droplet algorithm could be faster on GPU)
• Vetoing based on IPMs (available in smalldata_tools)
• Azimuthal integration by bins (for SAXS/WAXS)