New Support for CAMERA to Develop Computational Mathematics for Experimental Facilities Research

132 views Leave a comment

Experimental scholarship is evolving. With a appearance of new technology, systematic comforts are collecting information during augmenting rates and aloft resolution. However, creation clarity of this information is apropos a vital bottleneck. New arithmetic and algorithms are indispensable to remove useful information from these experiments.

To residence these flourishing needs, a Department of Energy has announced capitulation of a corner ASCR-BES-supported extend of $10.5 million over 3 years to enhance a Center for Advanced Mathematics for Energy Research Applications (CAMERA). CAMERA’s idea is to rise elemental arithmetic and algorithms, delivered as information research module that can accelerate systematic discovery. Originally focused on hurdles confronting a Advanced Light Source (ALS) and Molecular Foundry (MF) during a Lawrence Berkeley National Laboratory (Berkeley Lab), this new turn of support will assistance renovate CAMERA into a inhabitant apparatus opposite DOE computational, network, and light source facilities. ALS and MF are DOE Office of Science User Facilities.

Reaching opposite normal boundaries

“To tackle a rising hurdles during systematic user comforts opposite DOE, we need to invent and feat new arithmetic from a accumulation of opposite fields, and integrate that with initial imagination to aim systematic problems, ” says James Sethian, CAMERA Director, Head of Berkeley Lab’s Mathematics Group, and Professor of Mathematics during UC Berkeley.

CAMERA’s proceed is to arrange teams of practical mathematicians, statisticians, initial scientists, computational physicists, mechanism scientists and module engineers. By carrying a extended operation of imagination focused together in one place, CAMERA teams are means to strech opposite normal boundaries, find a common language, support initial information needs in mathematical terms and package mathematical solutions as module serviceable by a outmost community.

CAMERA launched in 2009 as a Laboratory Directed Research and Development (LDRD)program during Berkeley Lab. Based on this success, DOE’s Office of Advanced Scientific Computing Research (ASCR) and Basic Energy Science (BES) jointly invested in CAMERA as a commander project. This new turn of appropriation represents a vital enlargement in CAMERA’s idea and goals.

“The ultimate idea of a LDRD module is to capacitate bold, innovative projects with a intensity for high systematic impact. In this sense, CAMERA is a loyal success story,” says Horst Simon, Berkeley Lab Deputy Director. “Years ago, we saved CAMERA since we saw a potential. Since then, it has built on work achieved and upheld around DOE and has helped rise and broach mathematical research module opposite a operation of scholarship fields.”

CAMERA products embody new algorithms and module for ptychography (SHARP), extending occurrence shoal angle cat-scan pinch (HipGISAXS), reformation and research of imaged materials (QuantCT and F3D), chemical informatics for research of bright porous materials (Zeo++), quick methods for electronic structure calculations (PEXSI), nanocrystallography, and fluctuation X-ray scattering.

“With this new turn of funding, CAMERA is now staid to enhance on a initial efforts, tackle new mathematical hurdles that naturally emerge from elaborating initial needs, support visiting scientists and postdoctoral fellows from other Labs and facilities, and yield a curated module portal for modernized algorithms built by a village to accommodate these information challenges,” says David Brown, Director of Berkeley Lab’s Computational Research Division.

Tailoring Mathematics to Meet Experimental Needs

CAMERA tailors mathematics, algorithms and module privately to a initial mandate and computational resources during hand.

“With detectors being means to collect increasingly immeasurable amounts of information during accelerating rates, advances in computing architectures offer new opportunities to keep adult with a data-processing demands,” says Sethian, who is also a Co-Investigator of a Berkeley Institute of Data Sciences (BIDS). “And, rethinking a arithmetic mostly produces rare advances in correctness and speed, eccentric of computing power.”

Here are some examples of how CAMERA has tailored their proceed to opposite initial needs:

  • To accommodate in-situ real-time ptychographic imaging needs, CAMERA has recently embedded partial of a core rarely accelerated ptychography formula “SHARP” (Scalable Heterogeneous Adaptive Robust Ptychography) directly with CCD detector systems, producing real-time reformation of total diffraction and microscopy as a scanner moves over a sample.
  • To residence a picture research needs of micro-CT users “on a emporium floor,” CAMERA scientists have recently introduced Quant-CT and F3D, that are picture filtering and reformation algorithms formed on combinations of non-linear filters, PDE-based and template relating picture segmentation, and statistical sequence methods to investigate element porosity and topologies, deployed as user accessible Fiji plugins and regulating on workstations directly connected to ALS beamline 8.3.2.
  • As an instance of regulating remote supercomputing to accommodate information research needs, a SPOT Suite workflow government complement regulating during a National Energy Research Scientific Computing Center (NERSC) was used to emanate a antecedent information pipeline: as information was collected from an ALS GISAXS experiment, it was sent around DOE’s Energy Sciences Network (ESnet) to a Titan Supercomputer during a Oak Ridge Leadership Computing Facility (OLCF) for research on 8000 nodes regulating CAMERA’s HipGISAXS code, a customized high opening formula that exploits modernized graphics processors and molecule overflow optimization to fast retreat operative a representation from unnatural pinch patterns formed on twisted call Born approximations. NERSC and OLCF are DOE Office of Science User Facilities.
  • An instance of arithmetic essentially altering imaging capabilities comes from an rising imaging technique famous as fluctuation X-ray pinch (FXS). FXS offers a probability to boost a extractable information from resolution pinch experiments by several orders of bulk by exploiting absolute new light sources including DOE’s giveaway nucleus lasers, such as a Linac Coherent Light Source (LCLS) during SLAC National Accelerator Laboratory. However, a vital bottleneck has been a miss of techniques to establish ubiquitous molecular structure from a data. To residence this emanate and solve a reformation problem of FXS, CAMERA scientists introduced a absolute new mathematical speculation and algorithm, called multi-tiered iterative phasing (M-TIP). CAMERA’s new formula can fast establish ubiquitous structure in usually a few mins on a desktop computer.

CAMERA is jointly upheld by a ASCR and BES in a U.S. Department of Energy’s Office of Science. Further information about CAMERA, stream projects, destiny expansion, and rendezvous opportunities might be found during

Source: LBL