What is he best known for?
The fields of study he is best known for:
- Operating system
- Parallel computing
- Programming language
Samuel Williams mostly deals with Parallel computing, Multi-core processor, Stencil, Stencil code and Sparse matrix.
The various areas that Samuel Williams examines in his Parallel computing study include Sparse matrix-vector multiplication, Fast Fourier transform, Scalability and Programming paradigm.
His study on Scalability also encompasses disciplines like
- Data type which connect with Task parallelism,
- Computational complexity theory that connect with fields like Parallel algorithm.
His Programming paradigm research focuses on subjects like Supercomputer, which are linked to Intrinsics.
His studies in Multi-core processor integrate themes in fields like Computer architecture, Software portability, Memory hierarchy and Parallel processing.
His Parallel processing research includes themes of Software, Embedded system and Paradigm shift.
His most cited work include:
- The Landscape of Parallel Computing Research: A View from Berkeley (1818 citations)
- Roofline: an insightful visual performance model for multicore architectures (1340 citations)
- Stencil computation optimization and auto-tuning on state-of-the-art multicore architectures (456 citations)
What are the main themes of his work throughout his whole career to date?
Samuel Williams mainly investigates Parallel computing, Multi-core processor, Supercomputer, Stencil and Computational science.
His biological study deals with issues like Multigrid method, which deal with fields such as Benchmark.
His Multi-core processor research includes themes of Sparse matrix-vector multiplication, Software, Computation and Programming paradigm.
His studies deal with areas such as Concurrent computing, Cache, Microprocessor, Concurrency and Intrinsics as well as Supercomputer.
His Stencil research integrates issues from Optimizing compiler, Compiler, Memory hierarchy and Code generation.
His Compiler research includes elements of Thread, Embedded system and SIMD.
He most often published in these fields:
- Parallel computing (74.38%)
- Multi-core processor (32.23%)
- Supercomputer (16.53%)
What were the highlights of his more recent work (between 2016-2021)?
- Parallel computing (74.38%)
- Speedup (12.40%)
- Xeon Phi (10.74%)
In recent papers he was focusing on the following fields of study:
His scientific interests lie mostly in Parallel computing, Speedup, Xeon Phi, Benchmark and Stencil.
His Programming paradigm research extends to Parallel computing, which is thematically connected.
His Speedup study combines topics in areas such as Transpose, Eigenvalues and eigenvectors, LOBPCG and Matrix multiplication.
His Xeon Phi study incorporates themes from Lanczos resampling and Symmetric matrix.
In his study, Compiler, Memory hierarchy and Stencil code is inextricably linked to Code generation, which falls within the broad field of Stencil.
The study incorporates disciplines such as Porting, Multi-core processor, Data classification, Field-programmable gate array and Testbed in addition to Supercomputer.
Between 2016 and 2021, his most popular works were:
- AMReX: a framework for block-structured adaptive mesh refinement (56 citations)
- Hierarchical Roofline analysis for GPUs: Accelerating performance optimization for the NERSC‐9 Perlmutter system (15 citations)
- An Empirical Roofline Methodology for Quantitatively Assessing Performance Portability (15 citations)
In his most recent research, the most cited papers focused on:
- Operating system
- Programming language
- Parallel computing
Samuel Williams spends much of his time researching Parallel computing, Stencil, Code generation, Transpose and Software portability.
His studies link Program optimization with Parallel computing.
Samuel Williams combines subjects such as Compiler, Memory hierarchy and CUDA with his study of Stencil.
His studies in Memory hierarchy integrate themes in fields like Stencil code and Xeon.
His study in Transpose is interdisciplinary in nature, drawing from both Lanczos resampling, Thread, Symmetric matrix and Solver.
His Software portability research incorporates themes from Kernel, Computation, Computer engineering and FLOPS.
This overview was generated by a machine learning system which analysed the scientist’s
body of work. If you have any feedback, you can
contact us
here.
