Data Reduction Pipeline

Big-Picture To-do:

• implement algorithms (FPGA/GPU/Phi)
• implement supporting infrastructure (beam-center finding)
• validate effect of reduction on physics
• data transport
• ami support for setting drp parameters

Hutch Spreadsheets Showing Algorithms Used in 2015/2016

• amo/sxr (sioan)
• xpp/xcs (silke)
• DRP (cpo)

Possible Users To Contact

amo/sxr:

• nora berrah
• dan higley
• mankowsky
• urs staub (elsa abreu)
• marcus kubin (slow andor)
• emmanuel jal
• sam vinko
• christoph bostedt

xpp/xcs:

• timbvd
• mike kozina
• reis / mariano
• henrik
• ulvestad
• marco

SPI:

• Henry Chapman (Anton Barty, Kartik Ayyer, Andrew Morgan)
• Janos Hajdu (Filipe Maia)
• John Spence (Rick Kirian)
• Abbas Ourmazd (Peter Schwander)
• Veit Elser (Duane Loh, Ti-yen Lan)
• Jamie Sethian (Peter Zwart, Jeffrey Donatelli)

Crystallography:

• Ilme Schlichting (Thomas Barends, Lutz Foucar)
• Nick Sauter (Aaron Brewster)
• Henry Chapman (Anton Barty, Thomas White)
• Axel Brunger (Artem Lyubimov, Monarin Uervirojnangkoorn)
• Petra Fromme (Nadia Zatsepin)
• James Holton

Possible Simplifying Requirements

• don't change geometry
• if you do change geometry have a way to get new geometry info in a standard way
• stable beam-center (um, jets 50um). standard way to get beam center
• stable jets
• make jets/beam comparable unless you need stable to a pixel-size
• stable experimental setup
• don't switch between sparse/dense photon cases for spectrometer, and/or understand in advance
• measure detector tilts
• avoid double-beam-focus seen in SPI
• smaller pixels would mean photons don't occupy single-pixels, and we would get fooled less by hot-pixels
• gravity sensors on cameras
• more fiducial markings for geometry
• better calibrations for motor to positions
• hardware protection for cameras to avoid changes due to damage (camera replacement, bad-pixels)