Statistical evaluation of fracture of inclusions in cast aluminum alloy by massively-parallel voxel finite element analysis and geometrical measurements by Masaki TERANISHI, Osamu KUWAZURU, Masakazu KOBAYASHI, Hiroyuki TODA

“…The finite element analysis and its post-processing were performed on the supercomputers by the massively-parallel computing. The result of finite element analysis showed that the first principal stress was concentrated around the aligned or gathered Si particles, and the gradual increase by cyclic loading in the stress of the Si particles appeared near the outer surface and pores. …”
Get full text

A GPU-Accelerated Modern Fortran Version of the ECHO Code for Relativistic Magnetohydrodynamics by Luca Del Zanna, Simone Landi, Lorenzo Serafini, Matteo Bugli, Emanuele Papini

“…The chosen benchmark is the 3D propagation of a relativistic MHD Alfvén wave, for which strong and weak scaling tests performed on the LEONARDO pre-exascale supercomputer at CINECA are provided (using up to 256 nodes corresponding to 1024 GPUs, and over 14 billion cells). …”
Get full text

Spins of Supermassive Black Holes M87* and SgrA* Revealed from the Size of Dark Spots in Event Horizon Telescope Images by Vyacheslav Ivanovich Dokuchaev

“…The existence of hot accretion matter in the vicinity of black hole event horizons is predicted by the Blandford–Znajek mechanism, which is confirmed by recent general relativistic MHD simulations in supercomputers. A dark spot in the black hole image in the described model is a gravitationally lensed image of an event horizon globe. …”
Get full text

A Spatially Explicit Crop Yield Model to Simulate Agricultural Productivity for Past Societies under Changing Environmental Conditions by Maarten Van Loo, Gert Verstraeten

“…AquaCrop has been made spatially explicit, which allows hydrological interactions between different landscape positions, whilst computational time is kept limited by implementing parallelisation schemes on a supercomputer. The adapted model was calibrated and validated using crop and soil information sampled during the 2015 and 2016 harvest periods. …”
Get full text

Automated Cytogenetic Biodosimetry at Population-Scale by Peter K. Rogan, Eliseos J. Mucaki, Ben C. Shirley, Yanxin Li, Ruth C. Wilkins, Farrah Norton, Olga Sevriukova, Ngoc-Duy Pham, Ed Waller, Joan H. M. Knoll

“…ADCI-HT streamlines dose estimation using a supercomputer by optimal hierarchical scheduling of DC detection for varying numbers of samples and metaphase cell images in parallel on multiple processors. …”
Get full text

Structured Compression of Convolutional Neural Networks for Specialized Tasks by Freddy Gabbay, Benjamin Salomon, Gil Shomron

“…CNNs are increasingly deployed in environments from edge and Internet of Things (IoT) devices to high-end computational infrastructures, such as supercomputers, cloud computing, and data centers. The growing amount of data and the growth in their model size and computational complexity, however, introduce major computational challenges. …”
Get full text

Distributed Halide by Denniston, Tyler, Kamil, Shoaib, Amarasinghe, Saman P

“…The distributed benchmarks achieve up to 18× speedup on a 16 node testing machine and up to 57× speedup on 64 nodes of the NERSC Cori supercomputer.…”
Get full text
Get full text
Get full text

Advances in Bayesian Optimization with Applications in Aerospace Engineering by Poloczek, Matthias, Frazier, Peter, Lam, Remi Roger Alain Paul, Willcox, Karen E

“…The first method addresses optimization problems subject to inequality constraints where a finite budget of evaluations is available, a common situation when dealing with expensive models (e.g., a limited time to conduct the optimization study or limited access to a supercomputer). This challenge is addressed via a lookahead BO algorithm that plans the sequence of designs to evaluate in order to maximize the improvement achieved, not only at the next iteration, but once the total budget is consumed. …”
Get full text
Get full text
Get full text

TabulaROSA: Tabular Operating System Architecture for Massively Parallel Heterogeneous Compute Engines by Kepner, Jeremy, Brightwell, Ron, Edelman, Alan, Gadepally, Vijay N., Hayden, Jananthan, Jones, Michael, Madden, Samuel R., Michaleas, Peter W., Okhravi, Hamed, Pedretti, Kevin, Reuther, Albert I., Sterling, Thomas, Stonebraker, Michael

“…Simulations of forking in TabularROSA are performed using an associative array implementation and compared to Linux on a 32,000+ core supercomputer. Using over 262,000 forkers managing over 68,000,000,000 processes, the simulations show that TabulaROSA has the potential to perform operating system functions on a massively parallel scale. …”
Get full text

Deep Learning Emulators for Accessible Climate Projections by Lütjens, Björn

“…Climate models have been incredibly helpful in generating this causal understanding, but running them requires supercomputers and is only accessible to the minority of researchers. …”
Get full text
Get full text

Enabling technologies for self-aware adaptive systems by Santambrogio, Marco Domenico, Hoffmann, Henry Christian, Eastep, Jonathan Michael, Agarwal, Anant

“…Such a capability benefits a broad spectrum of computer systems from embedded systems to supercomputers and is particularly useful for meeting power, performance, and resource-metering challenges in mobile computing, cloud computing, multicore computing, adaptive and dynamic compilation environments, and parallel operating systems. …”
Get full text
Get full text

Memetic algorithm using multiple surrogates for complex engineering design optimization by Zhou, Zongzhao

“…Nevertheless, in analysis and design optimization processes where high-fidelity analysis codes are used, each exact fitness evaluation requiring the simulation of analysis codes may cost hours of supercomputer time. Therefore, the overwhelming part of the total run time in CED optimization is usually taken up by the simulation of analysis codes. …”
Get full text

Frustrated quantum magnets by Andrew Wibawa

“…While many computer hours have been spent on supercomputers on trying to understand the multistepped phenomena, much has yet to be revealed. …”
Get full text

Comparison between GPU and FPGA as hardware accelerator by Yang, Lu

“…The two common ways to speed it up are using GPU and FPGA as accelerators. The top supercomputer, such as Titan-Cray XK7 has employed GPUs as important parts of the system, which help the system achieve not only high performance but also high power efficiency. …”
Get full text

Heat equation analysis using parallel programming tools / Emilia Mohamad Shukree by Mohamad Shukree, Emilia

“…However, due to cost constrains, some organizations could not afford to have today's parallel supercomputers. Hence, urging developers and programmers to come up with a solution which is to virtualize the use of parallel processing which is why the Parallel Virtual Machine had come out as of today, to benefit organizations by connecting a few personal computers (PC’s) via today’s standard networking architectures. …”
Get full text

One-dimensional rabbit sinoatrial node models: benefits and limitations. by Garny, A, Kohl, P, Hunter, P, Boyett, MR, Noble, D

“…These one-dimensional models were implemented using CMISS on an SGI Origin 2000 supercomputer. Intercellular coupling parameters recorded in experimental studies on sinoatrial node and atrial cell-pairs under-represent the electrotonic interactions that any cardiomyocyte would have in a multidimensional setting. …”

Isolating aerosol-climate interactions in global kilometre-scale simulations by Herbert, R, Williams, A, Weiss, P, Watson-Parris, D, Dingley, E, Klocke, D, Stier, P

“…However, new state-of-the-art climate models running on exascale supercomputers are capable of representing these scales. …”

One-Dimensional Rabbit Sinoatrial Node Models: Benefits and Limitations by Garny, A, Kohl, P, Hunter, P, Boyett, MR, Noble, D

“…These one-dimensional models were implemented using CMISS on an SGI® Origin® 2000 supercomputer. Intercellular coupling parameters recorded in experimental studies on sinoatrial node and atrial cell-pairs under-represent the electrotonic interactions that any cardiomyocyte would have in a multidimensional setting. …”

Loop tiling in large-scale stencil codes at run-time with OPS by Reguly, I, Mudalige, G, Giles, M

“…We demonstrate strong and weak scalability up to 4608 cores of CINECA's Marconi supercomputer. We also evaluate our algorithms on Intel's Knights Landing, demonstrating maintained throughput as the problem size grows beyond 16GB, and we do scaling studies up to 8704 cores. …”

Adaptive energy efficient data center model by Mohd Nor, Norhashimi, Selamat, Mohd Hasan

“…Data centers had been evolving progressively since the first supercomputer invention. Although the main functions remain the same, the goals and characteristics change parallel with technology and market demands. …”
Get full text