Explore chapters and articles related to this topic
Proxy modeling
Published in Shahab D. Mohaghegh, Smart Proxy Modeling, 2023
“Originally, there was just experimental science, and then there was theoretical science, with Kepler's laws, Newton's laws of motion, Maxwell's equations, and so on. Then, for many problems, the theoretical models grew too complicated to solve analytically, and people had to start simulating. These simulations have carried us through much of the last half of the last millennium. At this point, these simulations are generating a whole lot of data, along with a huge increase in data from the experimental sciences. The world of science has changed, and there is no question about this. The techniques and technologies for such data-intensive science are so different that it is worth distinguishing data-intensive science from computational science as a new, fourth paradigm for scientific exploration” [Bell et al., 2009]. These were excerpts from a presentation by Jim Gray1 at the National Research Council where he described his vision of fourth paradigm of scientific research.
Parallel and high-performance systems
Published in Joseph D. Dumas, Computer Architecture, 2016
How can we classify systems based on modern GPU architectures? They are related to, but clearly not the same as, more conventional array processors that fall under the SIMD category of machines. It is generally agreed that GPUs, such as the AMD/ATI FireStream and NVIDIA’s Fermi and Kepler (as well as GPU-like processors, including the STI Cell Broadband Engine and Intel’s Xeon Phi), do not directly correspond to any of Flynn’s original four categories, so other terms have been proposed. NVIDIA, the leading provider of GPU hardware, refers to its GPU programming model in Flynn-like fashion as SIMT, or “Single Instruction Multiple Thread.” Others refer to the type of computing done by GPUs as “stream processing.” Both of these terms appear to be catching on among computing researchers, but what do they mean?
Accelerated parallel computation of field quantities for the boundary element method applied to stress analysis using multi-core CPUs, GPUs and FPGAs
Published in Cogent Engineering, 2018
Junjie Gu, Attila Michael Zsaki
The work presented in the paper considered two GPU-based accelerators: a desktop graphics card (GTX 760) and a purpose-built accelerator (NVIDIA Tesla K40c). Both of these were based on NIVIDIA’s Kepler microarchitecture (NVIDIA, 2013) and represented two common cards available at the time the research was conducted. The OpenCL-based acceleration of the BEM Algorithms 3 and 4 was used in the testing according to the test parameters and conditions set out in Sections 2.1 and 2.2.