Sparse matrix-vector multiplication (SpMV) is an important kernel in many scientific applications and known to be memory bandwidth limited. On modern processors with wide SIMD large numbers of cores, we identify address several bottlenecks which may limit performance even before bandwidth: (a) low efficiency due sparsity, (b) overhead irregular accesses, (c) load-imbalance non-uniform matrix structures.

Convolutional neural networks (CNNs) have become the critical technology to realize face detection and recognition in tracking (FT) system. However, traditional CNNs usually nontrivial computational time high energy consumption, making them inappropriate be deployed large-scale time-sensitive FT To address this challenge, we design an artificial intelligence Internet of Things (AIoT) empowered edge-cloud collaborative computing (ECCC) system based on energy-efficient field-programmable gate...

Protein complexes are crucial in improving our understanding of the mechanisms employed by proteins. Various computational algorithms have thus been proposed to detect protein from interaction networks. However, given massive interactome data obtained high-throughput technologies, existing algorithms, especially those with additionally consideration biological information proteins, either low efficiency performing their tasks or suffer limited effectiveness. For addressing this issue, work...

We conducted a micro benchmarking study of the time, energy, and power computation memory access on several existing platforms. These platforms represent candidate compute-node building blocks future high-performance computing systems. Our analysis uses "energy roofline" model, developed in prior work, which we extend two ways. First, improve model's accuracy by accounting for caps, basic hierarchy costs, measurement random patterns. Secondly, empirically evaluate server-, mini-,...

Named entity recognition, as a sub-task of information extraction, has attracted widespread attention from scholars at home and abroad since it was proposed, series studies discussions have been carried out based on it. This paper discusses the existing named recognition technology its history development.

Metal-organic framework (MOF) materials with tunable porous morphology, controlled crystalline structure, various compositions, and high specific surface area are widely used as precursors to synthesize electrocatalysts for water splitting, which is beneficial improving their oxygen evolution reaction (OER) performance. Using ZIF-67 a Co source Cu-BTC Cu source, hexagonal MOF-derived CuCoO2 (MOF-CCO) nanocrystals the size of ∼288 nm were prepared through one-step solvothermal method. The...

A new sparse high performance conjugate gradient benchmark (HPCG) has been recently released to address challenges in the design of linear solvers for next generation extreme-scale computing systems. Key computation, data access, and communication pattern HPCG represent building blocks commonly found today's HPC applications. While it is a well known challenge efficiently parallelize Gauss-Seidel smoother, most time-consuming kernel HPCG, our algorithmic architecture-aware optimizations...

Layered double hydroxides (LDHs) are an ultravioletlight (UV)-resistant material. In this study, two types of LDHs (Mg-Al-LDHs and Zn-Al-LDHs) were applied to modify bitumen by melt-blending. The effect ultraviolet aging on the rheology chemistry LDH-modified was studied means dynamic shear rheometer (DSR), thin-layer chromatography with flame ionization detection (TLC-FID), Fourier transform infrared spectroscopy (FTIR), Ultraviolet-Visible (UV-Vis) spectrophotometry reveal mechanisms...

This study examines the design of appointment scheduling policies with considerations not only conventional factors, for example, random consultation time and multiple patient types, but also a new factor, particularly, unpunctuality, that is, one may arrive earlier or later than time. Patient unpunctuality negatively affects system, such behaviour, reduces provider productivity clinical efficiency, increases health care costs, limits ability clinic to serve its patients population by...

In this work, nickel (Ni) doped Cu-BTC derived CuCoO2 (CCO) was successfully synthesized by a solvothermal method, and the effects of Ni doping concentration (such as 1 at%, 3 at% 5 at%) on crystal structure, morphology, composition oxygen evolution reaction (OER) catalytic performance were investigated. X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM), transmission (TEM) photoelectron spectroscopy (XPS) carried out to characterize morphology chemical crystals....

Hydrogen production from water splitting provides an effective method to alleviate the ever-growing global energy crisis. In this work, delafossite CuGaO2 (CGO) crystal was synthesized through hydrothermal routes with Cu(NO3)2·3H2O and Ga(NO3)3·xH2O used as reactants. The addition of cetyltrimethylammonium bromide (CTAB) found play important role in modifying morphology (CGO-CTAB). With CTAB, CGO-CTAB samples changed irregular flake typical hexagonal sheet microstructure, average size 1-2 μm...

The time and energy optimization of computationally intensive tasks involving unmanned air vehicles (UAVs) is highly important for increasing the reaction speed UAVs prolonging their lifetime. To achieve above objective, many studies based on heterogeneous computing have been carried out. Although these achieved good results, limitations remain. First, neural processing units (NPUs) emerged in recent years. However, insufficient attention has devoted to CPU/NPU research academia currently....

Layered double hydroxides (LDHs) are an ultraviolet-light (UV) resistant material. In this study, LDHs were used to modify bitumen. The optical and UV aging properties of modified bitumen investigated. Firstly, the thin films bitumen, with without LDHs, prepared. By using UV-Vis spectrophotometer, absorbance, reflectance, transmittance bituminous film evaluated. morphology LDHs-modified was observed by fluorescence microscopy (FM). Finally, resistance LDH-modified investigated UV-aging oven....

This paper presents a new optimized and scalable code for Hartree–Fock self-consistent field iterations. Goals of the design include scalability to large numbers nodes, capability simultaneously use CPUs Intel Xeon Phi coprocessors. Issues we encountered as scaled up on Tianhe-2 are described addressed. A major issue is load balance, which made challenging due integral screening. We describe general framework finding well-balanced static partitioning in presence Work stealing used polish...

This paper investigates a wireless system with multi-Unmanned Aerial Vehicles (UAVs) for improving the overall throughput. In contrast to previous studies that optimize locations of UAVs and channel assignment separately, this considers two issues jointly by exploiting Partially Overlapped Channels (POCs). The optimization problem maximizing network throughput is formulated as non-convex non-linear problem. order find practical solution, decomposed into subproblems, which are iteratively...

In this work, two different solvothermal synthesis routes were employed to prepare MOF-derived CuCoO2 (CCO) nanocrystals for electrocatalytic oxygen evolution reaction (OER) application. The effects of the reductants (ethylene glycol, methanol, ethanol, and isopropanol), NaOH addition, reactants, temperature on structure morphology product investigated. first route, Cu-BTC derived CCO (CCO1) with a size ∼214 nm specific surface area 4.93 m2 g-1 prepared by using Co(NO3)2·6H2O as Cu Co...

This study investigated the influence of layered double hydroxides (LDHs) on fatigue properties asphalt mixture. In this paper, different aging levels (thin film oven test (TFOT) and ultraviolet radiation (UV for short)) bitumen modified with various mass ratios LDHs were investigated. The TFOT UV process used to simulate short-term field thermal-oxidative long-term light bitumen, respectively. influences evaluated by dynamic shear rheometer (DSR) indirect tensile test. Results indicated...

Driven by the rapid technological advances, industrial Internet of Things (IIoT) has recently been embraced to enhance autonomous processes. Since a huge diverse traffic would be generated IIoT, processes meet challenges spectrum scarcity and on-demand service requirements. Millimeter wave (mmW) pervasive edge computing (PEC) technologies in 5G communication are available deal with these In this article, novel dual-band framework that integrates both mmW microwave (μW) networks PEC...

This paper presents the first sparse direct solver for distributed memory systems comprising hybrid multicourse CPU and Intel Xeon Pico-processors. It builds on algorithmic approach of SuperLU_DIST, which is right-looking statically pivoted. Our contribution a novel algorithm, called HALO. The name shorthand highly asynchronous lazy offload, it refers tithe way algorithm combines aggressive use asynchrony with accelerated updates, data shadowing (a la halo or ghost zones), all serve to hide...

Summary Energy optimization with time constraint has become a timely and significant challenge for the datacenters. In this paper, hardware software collaborative strategy is implemented to minimize energy cost while satisfying of aspect, DVFS‐capable CPU/GPU/FPGA heterogeneous computing infrastructure built. This can adjust its characteristics dynamically in terms run‐time contexts so that applications be executed efficiently less lower cost. deadline‐aware energy‐efficient task scheduling...

This paper presents optimizations in a high-performance conjugate gradient benchmark (HPCG) for multi-core Intel ® Xeon processors and many-core Phi™ coprocessors. Without careful optimization, the HPCG under-utilizes compute resources available modern due to its low arithmetic intensity challenges parallelizing Gauss–Seidel smoother (GS). Our optimized implementation fuses GS with sparse matrix vector multiplication (SpMV) address intensity, overcoming performance otherwise bound by memory...