Processing Unit Acceleration Research Articles

A Comprehensive Systematic Scoping Review of Self-Driving Vehicle Models

Self-driving vehicles (SDVs), also known as autonomous vehicles (AVs), are anticipated to revolutionize transportation by operating independently through the integration of machine learning algorithms, advanced processing units, and sensor networks. Numerous organizations globally are actively developing SDV models, prompting this paper’s objective to identify emerging trends and patterns in SDV development through a comprehensive systematic scoping review (SSR). This research involved selecting 85 relevant studies from an initial set of 551 records across multiple academic databases, utilizing well-defined inclusion and exclusion criteria along with snowballing techniques to ensure a thorough analysis. The findings emphasize critical technical specifications required for both full-scale and miniature SDV models, focusing on key software and hardware architectures, essential sensors, and primary suppliers. Additionally, the analysis explores publication trends, including publisher and venue distribution, authors’ affiliations, and the most active countries in SDV research. This work aims to guide researchers in designing their SDV models by identifying key challenges and exploring opportunities likely to shape future research and development in autonomous vehicle technology.

Performance analysis of amd ryzen 5 4600h mobile processor undervolting using AMD APU tuning utility on cinebench R23

In an effort to optimize laptop performance for gaming and high-demand applications without costly hardware upgrades, this research investigates the impact of CPU undervoltage using the AMD Ryzen Mobile 4600H processor. Undervolting, the process of reducing the CPU's voltage supply, is proposed as a strategy to enhance performance by lowering operational temperatures, potentially allowing for more efficient processing. This study uses the AMD APU Tuning Utility to adjust voltage settings and assesses performance changes using a series of benchmarks. Initial findings indicate that undervoltage can indeed have beneficial effects. The most significant data point from the research is the comparison of Cinebench R23 scores before and after applying undervolting settings. From a baseline score of 6835 points, system performance increased to 7880 points in the optimal undervolting scenario, an improvement of 1045 points. This shows a noticeable enhancement in processing efficiency. However, the study also reveals some complexities in undervolting, such as an initial drop in performance in the first configuration before gains are realized in subsequent adjustments. Efficiency values varied across different settings, starting with a decrease (-0.41) and culminating in a substantial gain (+1.54) by the fourth configuration. These results suggest that while undervolting can improve performance, the outcomes depend significantly on finding the right voltage balance, highlighting the nuanced nature of CPU voltage manipulation for performance optimization.

Environmental and economic implications of applying augmented reality in express parcel delivery in China

State-of-the-art information and communication technologies (ICTs), such as augmented reality (AR), have been increasingly adopted in express delivery. However, their environmental and economic impacts remain underexamined. We pioneer to assess the environmental and economic performances of applying AR in a parcel distribution center for express delivery, using a perspective life cycle assessment (LCA) approach and a traditional life cycle costing (LCC) method. On the basis of our field investigation and literature analysis, three scenarios, namely Baseline Scenario, AR Scenario, and AR+ Scenario, are developed consistent with current and potential AR applications in parcel delivery. Our findings indicate that AR and AR+ Scenarios are more environmentally sustainable than Baseline Scenario, exhibiting a decrease in global warming potential (GWP) by 13.44 % and 13.06 %, respectively. With the introduction of AR, the values of most selected impact categories show notable declines, except for human toxicity potential and marine ecotoxicity potential. The total endpoint scores decreased by 13.39 % and 12.86 % in the AR Scenario and AR+ Scenario, respectively. Moreover, the AR Scenario and AR+ Scenario are economically advantageous, demonstrating cost reductions of 38.59 % and 49.31 %, respectively, compared with the Baseline Scenario. Most substantial benefits associated with AR occur in the picking & sorting process, exhibiting improved accuracy and efficiency. Our sensitivity analysis underscores the substantial benefit associated with applying AR in routing improvement; 1 % routing improvement leads to a reduction of 0.27 kg CO2-Eq. Moreover, the sensitivity analysis also shows that labor cost is the most influential factor in the overall economic performance; its 1 % fluctuation leads to a change of 1.37 USD in the total cost. Our uncertainty analysis suggests that using more advanced processing units in AR devices leads to superior economic performance, but at greater environmental costs due to increased metallic contents. Based on these assessments, we provide managerial and theoretical implications on adopting smart ICTs in logistics.

Real-Time LEO Satellite Orbits Based on Batch Least-Squares Orbit Determination with Short-Term Orbit Prediction

The augmentation of the Global Navigation Satellite System (GNSS) by Low Earth Orbit (LEO) satellites is proposed as an effective method to improve the precision and shorten the convergence time of Precise Point Positioning (PPP). Serving as navigation satellites in the future, LEO satellites need to be provided with their high-accuracy orbits in real-time. This would potentially enable the high-accuracy real-time LEO satellite clock determination, and eventually facilitate the high-accuracy ground-based positioning. Studies have been performed to achieve such real-time orbits using a Kalman filter in both the kinematic and reduced-dynamic modes. Batch Least-Squares (BLS) adjustment delivers more stable orbits in near-real-time, as it performs better phase screening. However, it suffers from longer delays compared to the Kalman filter. With the LEO satellite orbit prediction strategies improved over time, this latency can be bridged by short-term orbit prediction. In this study, using real-time GNSS satellite products, the real-time LEO satellite orbits are obtained based on the batch least-squares adjustment and short-term prediction. LEO ephemeris parameters are generated within specific prediction time windows. Using real data from the 500 km GRACE C satellite and 810 km Sentinel-3B satellite, the near-real-time BLS Precise Orbit Determination (POD) results exhibit good accuracy with an Orbital User Range Error (OURE) of 2–4 cm using different real-time GNSS products. A range of delays of the BLS POD processes are assumed, based on tests performed on different processing machines, leading to various prediction windows, from 3–8 min to 12–17 min that correspond to the real-time usage. The orbital prediction errors are shown to be highly correlated with the orbital height and the prediction time. The computational efficiency thus becomes essential to reduce the prediction errors for a certain LEO satellite. For advanced processing units leading to a prediction window shorter or equal to 6–11 min, one can expect a total real-time orbital error budget of 3–5 cm, provided that an appropriate prediction strategy is applied and high-quality GNSS products are used. For a given fitting interval, the ephemeris fitting errors are generally related to the number of ephemeris parameters and the orbital height. Compared with the prediction errors, the ephemeris fitting errors do not play a significant role in the total error budget when using 22 ephemeris parameters.

Artificial intelligence-based classification with classical Turkish music makams: Possibilities to Turkish music education

Makams of Classical Turkish Music have been tried to be classified through various studies for the past years. Significant differences of opinion have emerged in the classification process of the makams in Music Education and Literacy from past to present. This situation creates problems in learning the makams related to music education and recognizing the makams heard. Additionally, there are uncertainties in the classification of the makam genre of the song, as individual mistakes were made while notating the musical notes. Apart from that, this situation constitutes a problem not only for the ones studying Turkish Classical Music but also for the ones interested in this certain type of Music. Therefore, the objective of the research is to contribute to the makam classification in Classical Turkish Music Education by developing an MIR system that determines the makam of the songs. Theoretically, we can extract the properties of sound signals with Time Wavelet Scattering Feature Extraction, classify them with SVM and distinguish between types of makams. In this study, upon eight different Makams, a Musical Information Retrieval system has been created via the Artificial Intelligence (AI) method of Support Vector Machines (SVM) and Time Wavelet Scattering Feature Extraction and through using a Graphics Processing Unit (GPU) accelerator for the sake of feature extraction. We performed the classification process by modeling it on the MATLAB program. The study's success rate was identified as 98.21% and it acquired a higher success rate compared to the other studies in the literature. After completing the classification procedure, the Makams were identified by sending samples belonging to different sound files from the system consisting of a database belonging to eight different Makams. In our study, the classification and detection processes were realized with nearly a hundred percent success. The difficulties encountered in classifying the makams in Classical Turkish Music mentioned above, with the application of artificial intelligence, the classification difficulty of individuals who have received this type of training or are interested in this subject has been overcome.

GPU Preconditioning for Block Linear Systems Using Block Incomplete Sparse Approximate Inverses

Solving sparse triangular systems is the building block for incomplete LU- (ILU-) based preconditioning, but parallel algorithms, such as the level-scheduling scheme, are sometimes limited by available parallelism extracted from the sparsity pattern. In this study, the block version of the incomplete sparse approximate inverses (ISAI) algorithm is studied, and the block-ISAI is considered for preconditioning by proposing an efficient algorithm and implementation on graphical processing unit (GPU) accelerators. Performance comparisons are carried out between the proposed algorithm and serial and parallel block triangular solvers from PETSc and cuSPARSE libraries. The experimental results show that GMRES (30) with the proposed block-ISAI preconditioning achieves accelerations 1.4 × –6.9 × speedups over that using the cuSPARSE library on NVIDIA Tesla V100 GPU.

Planning Advanced Treatment of Tap Water Consumption in Universitas Pertamina

The wastewater treatment plant (WWTP) in Universitas Pertamina’s area is operating very well. The existence of a green campus program with a wastewater recycling indicator has become one of the challenges. Improving the WWTP effluent quality by adding an advanced treatment unit. This study aims to design advanced processing units and estimate the effluent yields and required costs. This study was conducted by conducting a field survey, collecting water quality data, and literature study. With planning discharge (Qpeak) of 8.45 m3 /hour, the land area required for the addition of advanced treatment is 105.85 m2 . Advanced treatment consists of 1 unit of Equalization Tank, 2 units of slow sand filter, 1 unit of sand washer, 2 units of microfiltration membrane, 1 disinfection body, and 2 reservoir units. These units it is expected to make effluent quality meets the drinking water quality standards with a TSS value of 0 mg/L, Ammonia (NH3-N) 0.35 mg/L, Organic Substances (KmnO4) 0.513 mg/L, Total Dissolved Solids (TDS) 23 mg/L, and Total Coliform 0 Total/100 ml of sample. The total cost needed to build an advance treatment for tap water consumption is Rp 374,727,334.

Stereo Vision Based Pothole Detection System for Improved Ride Quality

<div class="section abstract"><div class="htmlview paragraph">Stereo vision based sensing systems have gained significant attention during the last two decades due to its reliable and accurate obstacle detection and recognition capabilities. Such systems with advanced processing units are now widely used in partially automated vehicles to improve passengers’ safety and comfort level. A predictive suspension control system that could provide better ride comfort and safety to the passengers by detecting potholes in advance and control the suspension system accordingly has been investigated in this study. Potholes can become serious safety hazard and can often cause discomfort if not detected and maneuvered at the right time. In this paper, a novel stereo vision based pothole detection system is proposed that detects pothole and calculates its depth accurately. In this proposed system, region of interest (ROI) of potential pothole candidates are selected utilizing intensity image and disparity image which is created using a pair of stereo images captured by a stereo camera. An intensity-depth based classifier has been developed which identifies the potholes from selected candidates and calculates its depth. Finally, 3D information of detected potholes is used to control the damping coefficient of the suspension system to improve the ride quality. The performance of the proposed pothole detection system has been evaluated using approximately 3.5 hours of driving video data captured with a frame rate of 20 frames/second. Experimental results show that, the accuracy of the proposed pothole detection system is about 84% and can detect pothole with ≥ 5 cm depth. Moreover, in-vehicle experiments confirm that the ride quality can be improved of about 16% utilizing the pothole detection system. The proposed system can be implemented for real-time applications in commercial vehicles and could provide significant benefits by improving safety and ride quality.</div></div>

Mechanochemical Treatment of Historical Tungsten Tailings: Leaching While Grinding for Tungsten Extraction Using NaOH

Innovative tungsten (W) extraction techniques are continually being sought because of challenges of low leaching efficiencies, despite using advanced processing units such as autoclaves operating high temperatures and pressures. Compared to conventional leaching, mechanochemical treatment improves the efficiency of leaching. Therefore, in this study, an innovative mechanochemical treatment method, referred to as leaching while grinding (LWG), was employed as a reprocessing option to optimize W recovery from historical tungsten tailings. Experiments were run using the regular two-level factorial design to screen through the four factors of stirrer speed, liquid/solid ratio, temperature, and digestion time to assess their criticality and effects in the LWG process. The stirrer speed and the liquid/solid ratio were the most critical factors in the optimization of W recovery. The maximum W recovery (91.2%) was attained at the highest stirrer speed (410 rpm), low liquid/solid ratio (0.8), long digestion time (6 h), and low leaching temperature (60 °C). The attained low leaching temperature (60 °C) was due to the mechanical activation of scheelite resulting from the simultaneous grinding and leaching. For such low- grade W material, liquid/solid ratio optimizing is critical for maintaining the digestion mixture fluidity, and for environmental and economic sustainability regarding the sodium hydroxide (NaOH) consumption, which was low.

APU Performance Evaluation for Accelerating Computationally Expensive Workloads

APUs (Accelerated Processing Units) are widely available in personal computers as low-cost processors that have a CPU and an integrated GPU for displaying graphics, in the same die. Using a ray tracing algorithm as a computationally intensive workload and taking advantage of the APU specific characteristics, we compare the performance of this SoC against CPU and GPU solutions in the same price range.

Thermal neutrons: a possible threat for supercomputer reliability

The high performance, high efficiency, and low cost of Commercial Off-The-Shelf (COTS) devices make them attractive for applications with strict reliability constraints. Today, COTS devices are adopted in HPC and safety-critical applications such as autonomous driving. Unfortunately, the cheap natural boron widely used in COTS chip manufacturing process makes them highly susceptible to thermal (low energy) neutrons. In this paper, we demonstrate that thermal neutrons are a significant threat to COTS device reliability. For our study, we consider two DDR memories, an AMD APU, three NVIDIA GPUs, an Intel accelerator, and an FPGA executing a relevant set of algorithms. We consider different scenarios that impact the thermal neutron flux such as weather, concrete walls and floors, and HPC liquid cooling systems. Correlating beam experiments and neutron detector data, we show that thermal neutrons FIT rate could be comparable or even higher than the high energy neutron FIT rate.

Bonsai-SPH: A GPU accelerated astrophysical Smoothed Particle Hydrodynamics code

We present the smoothed-particle hydrodynamics simulation code, Bonsai-SPH, which is a continuation of our previously developed gravity-only hierarchical NN-body code (called Bonsai). The code is optimized for Graphics Processing Unit (GPU) accelerators which enables researchers to take advantage of these powerful computational resources. Bonsai-SPH produces simulation results comparable with state-of-the-art, CPU based, codes, but using an order of magnitude less computation time. The code is freely available online and the details are described in this work.

OpenCL-Darknet: implementation and optimization of OpenCL-based deep learning object detection framework

Object detection is a technology that deals with recognizing classes of objects and their location. It is used in many different areas, such as in face-detecting systems [16, 34, 37], surveillance tools [9], human-machine interfaces [17], and self-driving cars [18, 23, 25, 26, 30]. These days, deep learning object detection approaches have achieved significantly better performance than the classical feature-based algorithms. Darknet [31] is a deep learning object detection framework, which is well known for its fast speed and simple structure. Unfortunately, Darknet can only work with Nvidia CUDA [6] for accelerating its deep learning calculations. For this reason, users have only limited options of selecting appropriate graphic cards. Open computing language (OpenCL) [35], an open standard for cross-platform, parallel programming of heterogeneous systems, is available for the general hardware accelerators. However, many deep learning frameworks including Darknet have no support for OpenCL. In our previous paper, we presented OpenCL-Darknet [19], which transformed the CUDA-based Darknet into an open standard OpenCL backend. The original OpenCL-Darknet successfully showed its ability for the general graphics processing unit (GPU) hardware. However, it could not achieve competitive performance compared with the CUDA version, and it only supported a limited platform. In this study, we improved the performance of OpenCL-Darknet with several optimization techniques and added support for various architectures. We also evaluated OpenCL-Darknet not only in AMD R7 accelerated processing unit (APU) with OpenCL 2.0, but also in Nvidia GPU and ARM Mali embedded GPU with OpenCL 1.2 Profile. The evaluation using the standard object detection datasets showed that our advanced OpenCL-Darknet reduced the processing time by at most 50% on average for various deep learning object detection networks compared with our original implementation. We also showed that our OpenCL deep learning framework has competitiveness compared with the CUDA-based one.

High-Energy Versus Thermal Neutron Contribution to Processor and Memory Error Rates

We present the results of accelerated radiation testing on an AMD accelerated processing unit, three Nvidia graphic processing units, an Intel accelerator, a field-programmable gate array, and two double-data-rate memories under thermal and high-energy neutrons separately. The sensitivity depends on the device type and the code being executed and we show that thermal neutrons contribute to the error rate of modern computing devices under certain conditions.

Sn-3.0Ag-0.5Cu/Sn-58Bi composite solder joint assembled using a low-temperature reflow process for PoP technology

Package-on-Package (PoP) is a popular technology for fabricating chipsets of accelerated processing units. However, the coefficient of thermal expansion mismatch between Si chips and polymer substrates induces thermal warpage during the reflow process. As such, the reflow temperature and reliability of solder joints are critical aspects of PoP. Although Sn58Bi is a good candidate for low-temperature processes, its brittleness causes other reliability issues. In this study, an in-situ observation was performed on composite solders (CSs) made of an Sn-3.0Ag-0.5Cu (SAC305) solder and an Sn58Bi solder that were mixed during reflow at temperatures of 170 °C (CS-170), 180 °C (CS-180), and 190 °C (CS-190). The volumes of the mixed-solder regions were in the order of CS-190 > CS-180 > CS-170. A calculated Sn58Bi-SAC305 isoplethal section of the Sn-Ag-Cu-Bi quaternary system was employed to elucidate the melting behavior of CS-190. CS-190 with Sn cyclic twin boundary and Bi phase preferred orientation of [0001] was demonstrated by electron backscatter diffraction. Moreover, Cu/CS-190/Cu joints with a shear strength of 46 MPa were formed. CS-190 can reduce the brittleness of SnBi solder owing to the reduced solid-solution hardening with the decrease in the Bi content. These factors can improve the reliability of PoP technology.

Dual tree traversal on integrated GPUs for astrophysical N-body simulations

In astrophysical N-body simulations, O( N) fast multipole methods (FMMs) with dual tree traversal (DTT) on multi-core CPUs are faster than O( N log N) CPU tree-codes but can still be outperformed by GPU ones. In this article, we aim at combining the best algorithm, namely FMM with DTT, with the most powerful hardware currently available, namely GPUs. In the astrophysical context requiring low accuracies and non-uniform particle distributions, we show that such combination can be achieved thanks to a hybrid CPU-GPU algorithm on integrated GPUs: while the DTT is performed on the CPU cores, the far- and near-field computations are all performed on the GPU cores. We show how to efficiently expose the interactions resulting from the DTT to the GPU cores, how to deploy both the far- and near-field computations on GPU, and how to overlap the parallel DTT on CPU with GPU computations. Based on the falcON code and using OpenCL on AMD Accelerated Processing Units and on Intel integrated GPUs, this first heterogeneous deployment of DTT for FMM outperforms standard multi-core CPUs and matches GPU and high-end CPU performance, being hence more cost- and power-efficient.

ASW: Accelerating Smith–Waterman Algorithm on Coupled CPU–GPU Architecture

Smith–Waterman algorithm (SW) is a popular dynamic programming algorithm widely used in bioinformatics for local biological sequence alignment. Due to the $$O(n^2)$$ high time and space complexity of SW and growing size of biological data, it is crucial to accelerate SW for high performance. In view of the GPU high efficiency in science computation, many existing studies (e.g., CUDAlign, CUDASW++) speedup SW with GPU. However, the strong data dependency makes SW communication intensive, and the previous works fail to fully leverage the heterogeneous capabilities of the GPU machine for either the neglect of the CPU ability or the low bandwidth of PCI-e. In this paper, we propose ASW, which aims at accelerating SW algorithm with accelerated processing unit (APU), a heterogeneous processor integrates CPU and GPU in a single die and share the same memory. This coupled CPU–GPU architecture is more suitable for frequent data exchanging due to the elimination of PCI-e bus. For the full utilization of both CPU and GPU in APU system, ASW partitions the whole SW matrix into blocks and dynamically dispatches each block to CPU and GPU for the concurrent execution. A DAG-based dynamic scheduling method is presented to dispatch the workload automatically. Moreover, we also design a time cost model to determine the partition granularity in the matrix division phase. We have evaluated ASW on AMD A12 platform and our results show that ASW achieves a good performance of 7.2 GCUPS (gigacells update per second).

Toward a software transactional memory for heterogeneous CPU–GPU processors

The heterogeneous accelerated processing units (APUs) integrate a multi-core CPU and a GPU within the same chip. Modern APUs implement CPU–GPU platform atomics for simple data types. However, ensuring atomicity for complex data types is a task delegated to programmers. Transactional memory (TM) is an optimistic approach to achieve this goal. With TM, shared data can be accessed by multiple computing threads speculatively, but changes are only visible if a transaction ends with no conflict with others in its memory accesses. In this paper we present APUTM, a software TM designed for APU processors which focuses on minimizing the access to shared metadata. The main goal of APUTM is to understand the trade-offs of implementing a software TM on such platform. In our experiments, APUTM is able to outperform sequential execution of the applications. Additionally, we compare its adaptability to execute in one of the devices or in both simultaneously.

Energy-Delay-FIT Product to compare processors and algorithm implementations

We propose an extension to the Energy-Delay-Product (EDP) metric to compare different processors considering not only their energy consumption and execution time but also reliability. The Energy-Delay-FIT Product (EDFP) allows a pragmatic evaluation of the most suitable device to run an application.We consider three representative benchmarks and apply EDFP to compare Intel Xeon-Phi co-processors, NVIDIA K40 Graphics Processing Units (GPUs), and AMD Kaveri Accelerated Processing Units (APUs). Our results show that HPC processors have higher power consumption and are more prone to be corrupted than APUs. However, the overall trade-off is attenuated by HPC processors efficiency, which makes them the most suitable candidates for the great majority of the considered applications.Additionally, we use EDFP to compare optimized and naive implementations of three benchmarks as executed on NVIDIA GPUs. Our results show that the naive implementation has generally better EDFP only for small input sizes while the optimized implementations are more efficient and reliable once the GPU resources are saturated.

APLIKASI MEMBRANE BIOREACTOR (MBR) UNTUK PROSES DAUR ULANG AIR LIMBAH

The population is growing rapidly result to increased demand for drinking water while the other side of this situation causes to the increasing amount of wastewater. Waste water treatment plant has been built in several major cities in Indonesia to overcome the problem of pollution. Wastewater treatment plant only treats wastewater to tackle environmental pollution without being able to overcome the shortage of clean water community. Therefore, to solve the existing problems will be considered any further processing with recycled waste water treatment. One technology that may be developed for this purpose is the system of Membrane Bioreactor (MBR). This has been widely applied as an advanced unit of wastewater treatment and serves as a unit of processing recycled waste water. The MBR system has advantages such as: it only takes a little bit of land, can treat wastewater with discharge varying quality and process wastewater with high pollutant removal efficiency. MBR were investigated using a membrane filter submerged in the reactor. Membrane module is hollow fiber with a form of polymeric membrane materials. Research has been done on wastewater treatment with a combination of conventional WWTP as preliminary and advanced processing unit with MBR since March 4, 2014 until June 2014. The effluent quality of result research proved that the treated water quality meets water quality standards are set with Permenkes RI N0.416/Menkes/Per/IX/1990, dated 3 September 1990. The nitrification and denitrification also proceed well so that the concentration of nitrite and nitrate meet the quality standards. In addition, the percentage removal of heavy metals (arsenic, cadmium, chromium, selenium, lead) can reach 99%. Keywords: MBR, wastewater reuse, BOD5, COD, HRT dan SRT