big data

On The Coupling between HPC and Statistics: Challenges, Opportunities, and Future Trends of Emerging Techniques -- (Virtual SIAM CSE21 Minisymposium)

-

KAUST

Abstract Big data modeling/inference and large-scale simulations have followed largely independent paths to the high-performance computing (HPC) frontier, but important opportunities now arise that can be addressed by combining the strengths of each. HPC is becoming increasingly significant in scaling existing statistical methods to larger and more complex applications and developing novel methods that are amenable to scaling within the constraints that exist in modern HPC architectures. The purpose of this minisymposium is to bring together researchers in the area of statistics and HPC to

Keep the data coming

1 min read ·
A continuous data supply ensures data-intensive simulations can run at maximum speed.

Mixing precision for model acceleration

1 min read ·
A mixed-precision approach for modeling large geospatial datasets can achieve benchmark accuracy with a fraction of the computational run time.

Code of collaboration

1 min read ·
David Keyes's and Matteo Ravasi's cross-disciplinary project results in improved efficiency for seismic processing, with promising applications for the energy industry.

Kadir Akbudak

Dr. Kadir Akbudak is a Consultant at Extreme Computing Research Center at KAUST, working with Professor David E. Keyes and Dr. Hatem Ltaief. Research Interests Kadir Akbudak's research interests include Combinatorial scientific computing, Graph and hypergraph partitioning, Locality-aware partitioning and scheduling, High performance kernels for sparse computations, Vectorization for irregular applications, Iterative numerical methods and Parallel numerical algorithms. Education Profile Ph.D., Computer Engineering, Bilkent University, 2015. M.S., Computer Engineering, Bilkent University, 2009