Multi-core Research Articles

ENHANCING GAME ENGINE PERFORMANCE THROUGH MULTITHREADING

Today, game engines play a significant role in the modern gaming world, with possible multithreading techniques being pivotal in optimizing gaming performance and overcoming performance bottlenecks. This paper discusses a worthwhile approach to increasing game engine performance by leveraging effective multithreading principles, along with the issues that need to be considered in the design of a multithreaded game engine and aims to contribute to the continuing evolution of game development. Through an ideal integration of data and task parallelisms in a well-balanced engine, developers can fully benefit from the capabilities of multicore processors, resulting in smoother and more responsive gaming experiences. The results obtained from employing both data and task parallelism jointly in the same architecture showed a remarkable performance increase of about 2.3 times faster than the single-threaded model. The paper also addresses common multithreading challenges, such as synchronization issues and load balancing, together with their solutions, emphasizing the effectiveness of the proposed model. Keywords: Multithreaded Game Engine, Multithreading, Game Engine Performance, Multithreading Principles, Data Parallelism, Task Parallelism, Task scheduler.

Improvement and Application of Directional Reaming While Willing Bit in Tahe Oilfield

Unlike conventional sidetracking technology, the technology of reaming while directional drilling makes it possible for us to conduct directional drilling and reaming simultaneously using dual core bit, which simplifies the construction steps and can greatly shorten the drilling cycle and reduce drilling costs. With the increasing of the tasks of sidetrack horizontal well in the old wells of Tahe Oilfield, the technology of reaming while directional drilling has become an important means for sidetrack of old wells in Tahe Oilfield. However the substratum of Bachu formation has high content of gravel with large diameter and strong consolidation, which cause severe abrasion of dual core bit, short drill life and low ROP. With the problems in the substratum of Bachu Formation, Tahe Oilfield continuously improves dual core bit based on its effect of reaming while drilling, and breaks the national records of the maxim setting depth, inclination and continuous expansion length during the operations of well TH12124CH. This paper shows the improvements of dual core bit and its application effect both in well TH10233CH and well TH12124CH, and put forward a further scheme to improve the dual core bit for the substratum of Bachu Formation of Tahe Oilfield.

A cyclic-track decomposition method for 3D MOC neutron transport simulation

The method of characteristics (MOC) is a promising numerical technique for solving the neutron transport equation. This method has been parallelized to accommodate its high computing power and memory requirements. Angle decomposition is widely used by parallel MOC codes to distribute MOC tracks over CPU cores, but it may lead to load imbalance. Based on modular ray tracing, we present a cyclic-track decomposition method along with a load balancing strategy to address the limitations of angle decomposition for parallel MOC on 3D axially extruded geometries. It has been implemented as the top-level decomposition in the ANT-MOC code to demonstrate its effectiveness and to show its potential as an alternative to domain decomposition. We evaluated the cyclic-track decomposition method using a generalized performance model and some neutron transport benchmarks. Our evaluation showed that the cyclic-track decomposition method generated a more balanced load distribution than the angle decomposition method, and ANT-MOC with the new strategy for load balancing achieved a good performance of over 40% parallel efficiency as it scales from 10,240 to 98,304 cores on the ORISE supercomputer.

A novel parallel heuristic method to design a sustainable medical waste management system

Efficient management of waste generated in healthcare systems is crucial to minimize its environmental impact and ensure public health. Sustainable medical waste management (MWM) systems require careful network design, which can be achieved through efficient optimization techniques. This work develops a mixed-integer linear programming (MILP) to formulate the problem, a two-step MILP (TSMILP) to generate quality lower bounds, and a novel parallel heuristic algorithm to configure a sustainable waste management system including waste generation centers (WGCs), waste treatment centers (WTCs), waste recycling centers (WRCs), waste disposal centers (WDCs) and waste incineration centers (WICs). Such a hybrid methodology has not been yet offered in the literature wherein the aim is to address strategic (establishment of facilities), tactical (employment of transportation system), and operational decisions (transportation planning) optimally in large networks. As reflected in the literature, there is a huge gap in efficiency and application of combinatorial optimization, and parallel computing in sustainable MWM systems, where the suggested MILPs' solvers are not technically capable of discovering quality solutions in reasonable runtimes on large-sized instances. Thus, we suggest a novel heuristic equipped with parallel computing to share the complexity of the problem, with all the CPU cores to shorten runtime. Comparing the results generated by the parallel heuristic with those of the sequential heuristic, the MILP, and the TSMILP on three sets of benchmark instances using Nemenyi's post-hoc procedure for Friedman's test, it is inferred that the parallel heuristic is so effective in coping with the problem, and produces high-quality solutions, especially on the large-sized set. Finally, sensitivity analysis is adopted to analyze the effects of parameters on the objective values and provide useful managerial insights.

Real-time enhanced efficient thread level parallelism scheme for performance improvement in heterogeneous edge computing

In the era of technology, there is a need to rely on new high performance Heterogeneous embedded computing device to process a huge amount of data for various smart applications. Packing different architecture processor into a system on chip provides a substantial potential improvement in computing horsepower, but the maximum processing power of this heterogeneous edge computing processor can only be harnessed if the target software is actually configured to utilize all the processing elements. The proposed Enhanced Efficient thread level parallelism (EETLP) is implemented using CUDA in CPU-GPU based heterogeneous edge computing platform and analyzed with different size of matrix multiplication. From the experiment results, it was clearly observed that for the matrix size 1024x1024, Efficient Thread Level Parallelism (ETLP) using quad core CPU processor reduces 71% execution time and EETLP reduces 99% execution time compared to Basic Sequential Execution (BSE). In terms of Speedup, EETLP has achieved 5.5Kx speedup compare with ETLP and 19Kx speedup against BSE on CPU.

An approach for low-power heterogeneous parallel implementation of ALC-PSO algorithm using OmpSs and CUDA

PSO (particle swarm optimization), is an intelligent search method for finding the best solution according to population state. Various parallel implementations of this algorithm have been presented for intensive-computing applications. The ALC-PSO algorithm (PSO with an aging leader and challengers) is an improved population-based procedure that increases convergence rapidity, compared to the traditional PSO. In this paper, we propose a low-power heterogeneous parallel implementation of ALC-PSO algorithm using OmpSs and CUDA, for execution on both CPU and GPU cores. This is the first effort to heterogeneous parallel implementing ALC-PSO algorithm with combination of OmpSs and CUDA. This hybrid parallel programming approach increases the performance and efficiency of the intensive-computing applications. The proposed approach of this article is also useful and applicable for heterogeneous parallel execution of the other improved versions of PSO algorithm, on both CPUs and GPUs. The results demonstrate that the proposed approach provides higher performance, in terms of delay and power consumption, than the existence implementations of ALC-PSO algorithm.

MobileInst: Video Instance Segmentation on the Mobile

Video instance segmentation on mobile devices is an important yet very challenging edge AI problem. It mainly suffers from (1) heavy computation and memory costs for frame-by-frame pixel-level instance perception and (2) complicated heuristics for tracking objects. To address these issues, we present MobileInst, a lightweight and mobile-friendly framework for video instance segmentation on mobile devices. Firstly, MobileInst adopts a mobile vision transformer to extract multi-level semantic features and presents an efficient query-based dual-transformer instance decoder for mask kernels and a semantic-enhanced mask decoder to generate instance segmentation per frame. Secondly, MobileInst exploits simple yet effective kernel reuse and kernel association to track objects for video instance segmentation. Further, we propose temporal query passing to enhance the tracking ability for kernels. We conduct experiments on COCO and YouTube-VIS datasets to demonstrate the superiority of MobileInst and evaluate the inference latency on one single CPU core of the Snapdragon 778G Mobile Platform, without other methods of acceleration. On the COCO dataset, MobileInst achieves 31.2 mask AP and 433 ms on the mobile CPU, which reduces the latency by 50% compared to the previous SOTA. For video instance segmentation, MobileInst achieves 35.0 AP and 30.1 AP on YouTube-VIS 2019 & 2021.

Architectural and Technological Approaches for Efficient Energy Management in Multicore Processors

Benchmarks play an essential role in the performance evaluation of novel research concepts. Their effectiveness diminishes if they fail to exploit the available hardware of the evaluated microprocessor or, more broadly, if they are not consistent in comparing various systems. An empirical analysis of the consecrated Splash-2 benchmarks suite vs. the latest version Splash-4 was performed. It was shown that on a 64-core configuration, half of the simulated benchmarks reach temperatures well beyond the critical threshold of 105 °C, emphasizing the necessity of a multi-objective evaluation from at least the following perspectives: energy consumption, performance, chip temperature, and integration area. During the analysis, it was observed that the cores spend a large amount of time in the idle state, around 45% on average in some configurations. This can be exploited by implementing a predictive dynamic voltage and frequency scaling (DVFS) technique called the Simple Core State Predictor (SCSP) to enhance the Intel Nehalem architecture and to simulate it using Sniper. The aim was to decrease the overall energy consumption by reducing power consumption at core level while maintaining the same performance. More than that, the SCSP technique, which operates with core-level abstract information, was applied in parallel with a Value Predictor (VP) or a Dynamic Instruction Reuse (DIR) technique, which rely on instruction-level information. Using the SCSP alone, a 9.95% reduction in power consumption and an energy reduction of 10.54% were achieved, maintaining the performance. By combining the SCSP with the VP technique, a performance increase of 8.87% was obtained while reducing power and energy consumption by 3.13% and 8.48%, respectively.

PARamrfinder: detecting allele-specific DNA methylation on multicore clusters

The discovery of Allele-Specific Methylation (ASM) is an important research field in biology as it regulates genomic imprinting, which has been identified as the cause of some genetic diseases. Nevertheless, the high computational cost of the bioinformatic tools developed for this purpose prevents their application to large-scale datasets. Hence, much faster tools are required to further progress in this research field. In this work we present PARamrfinder, a parallel tool that applies a statistical model to identify ASM in data from high-throughput short-read bisulfite sequencing. It is based on the state-of-the-art sequential tool amrfinder, which is able to detect ASM at regional level from Bisulfite Sequencing (BS-Seq) experiments in the absence of Single Nucleotide Polymorphism information. PARamrfinder provides the same Allelically Methylated Regions as amrfinder but at significantly reduced runtime thanks to exploiting the compute capabilities of common multicore CPU clusters and MPI RMA operations to attain an efficient dynamic workload balance. As an example, our tool is up to 567 times faster for real data experiments on a cluster with 8 nodes, each one containing two 16-core processors. The source code of PARamrfinder, as well as a reference manual, is available at https://github.com/UDC-GAC/PARamrfinder.

Spatial-temporal evolution and influencing factors of ecological resilience in urban agglomerations: a case study of Shanxi section of the Yellow River Basin

Scientific assessment of the development status and factors influencing the urban ecological resilience of the Yellow River Basin (YRB) is highly significant for promoting its development. This study constructed an evaluation index system for urban ecological resilience considering the four dimensions of pressure, state, response, and innovation. The spatiotemporal ecological resilience of the urban agglomeration (UA) in the Shanxi section of the YRB from 2012 to 2021 was studied using kernel density estimation, Dagum Gini coefficient, and standard deviation ellipse, and the influencing factors of urban ecological resilience were analyzed using a geographic detector. This research revealed that (1) the ecological resilience of cities in the Shanxi section of the YRB experienced a fluctuation process of rise—fall—rise. The urban ecological resilience generally reflected a gradient decreasing spatial pattern of Central Shanxi UA > South Shanxi UA > North Shanxi UA, and gradually changed from the dual core of “Taiyuan—Jincheng City” to the single core of Taiyuan City. (2) The migration trajectory of urban ecological resilience center of gravity fluctuated in the direction of “northwest-southeast,” and moved 12.63 km to the southeast overall. (3) The water supply per ten thousand Yuan GDP, occupied area of construction land per ten thousand Yuan GDP, green coverage rate in built-up areas, ratio of research and experimental development funds (R&D) to GDP, proportion of science and technology expenditure in local fiscal expenditure, and patent licensing quantity index have a high influence on urban ecological toughness at all stages. This influence was further strengthened by the interaction between factors. This study provides an important scientific basis for shaping high-quality development advantages in the YRB and creating a resilient and livable environment.

Automated Symbolic Processes for Dynamic Modeling of Redundant Manipulator Robots

In this study, groundbreaking software has been developed to automate the generation of equations of motion for manipulator robots with varying configurations and degrees of freedom (DoF). The implementation of three algorithms rooted in the Lagrange–Euler (L-E) formulation is achieved through the utilization of .m files in MATLAB R2020a software.This results in the derivation of a symbolic dynamic model for industrial manipulator robots. To comprehend the unique features and advantages of the developed software, dynamic simulations are conducted for two 6- and 9-DoF redundant manipulator robots as well as for a 3-DoF non-redundant manipulator robot equipped with prismatic and rotational joints, which is used to simplify the dynamic equations of the redundant prototypes. Notably, for the 6-DoF manipulator robot, model predictive control (MPC) is employed using insights gained from the dynamic model. This enables optimal control by predicting the future evolution of state variables: specifically, the values of the robot’s joint variables. The software is executed to model the dynamics of different types of robots, and the CPU time for a MacBook Pro with a 3 GHz Dual-Core Intel Core i7 processor is less than a minute. Ultimately, the theoretical findings are validated through response graphs and performance indicators of the MPC, affirming the accurate functionality of the developed software. The significance of this work lies in the automation of motion equation generation for manipulator robots, paving the way for enhanced control strategies and facilitating advancements in the field of robotics.

DoppelGanger++: Towards Fast Dependency Graph Generation for Database Replay

A database replay system (DRS) captures workloads on a production system and then replays them in a test system to test various system changes, avoiding any risk before realizing them in production. The dependency graph generation in a DRS is crucial in preserving output determinism while maximizing concurrency. The state-of-the-art dependency graph generation algorithm deployed in a commercial DBMS uses a generate-and-prune strategy. It first generates a dependency graph by performing backward scans for each request in a workload. It then prunes all redundant edges using an expensive, transitive reduction algorithm. However, we notice that this generates a large dependency graph that contains many redundant edges and its worst-case time complexity is quadratic to the number of requests in a workload. In order to solve these challenging problems, we formally propose four classes of dependency graphs for DRSs. We then present a stateful single forward scan algorithm, SSFS, to generate any class of dependency graphs by performing a single scan over all requests while succinctly maintaining states. Here, states refer to information that is stored and maintained for efficient dependency graph generation. We also propose the parallel SSFS to utilize the computation power with multi-core CPUs while balancing the loads. We implemented our DRS in a leading commercial DBMS. Extensive experiments using the TPC-C, SD benchmarks, and a real-world customer workload show that our DRS significantly improves the dependency graph generation time by up to two orders of magnitude, compared to the state-of-the-art.

Crosstalk Analysis of Few Mode Multi Core Fiber Using Selective Mode Launch

Crosstalk Analysis of Few Mode Multi Core Fiber Using Selective Mode Launch

Optimizing Large Language Models on Multi-Core CPUs: A Case Study of the BERT Model

The BERT model is regarded as the cornerstone of various pre-trained large language models that have achieved promising results in recent years. This article investigates how to optimize the BERT model in terms of fine-tuning speed and prediction accuracy, aiming to accelerate the execution of the BERT model on a multi-core processor and improve its prediction accuracy in typical downstream natural language processing tasks. Our contributions are two-fold. First, we port and parallelize the fine-tuning training of the BERT model on a multi-core shared-memory processor. We port the BERT model onto a multi-core processor platform to accelerate the fine-tuning training process of the model for downstream tasks. Second, we improve the prediction performance of typical downstream natural language processing tasks through fine-tuning the model parameters. We select five typical downstream natural language processing tasks (CoLA, SST-2, MRPC, RTE, and WNLI) and perform optimization on the multi-core platform, taking the hyperparameters of batch size, learning rate, and training epochs into account. Our experimental results show that, by increasing the number of CPUs and the number of threads, the model training time can be significantly reduced. We observe that the reduced time is primarily concentrated in the self-attention mechanism. Our further experimental results show that setting reasonable hyperparameters can improve the accuracy of the BERT model when applied to downstream tasks and that appropriately increasing the batch size under conditions of sufficient computing resources can significantly reduce training time.

Spatial spillover and threshold effects of digital rural development on agricultural circular economy growth

The infusion of digital technologies imparts impetus to the development of agricultural circular economy. Study focuses on the nexus of digital rural development, agricultural circular economy growth, employing entropy, Critic, and TOPSIS models to calculate the evolution of agricultural circular economy from 2011 to 2021. Furthermore, the temporal and spatial dynamics of agricultural circular economy development are explored. In addition, spatial Durbin models and threshold models are utilized to explore the spatial spillover effects of digital rural development on agricultural circular economy, and the threshold characteristics of agricultural circular economy with varying levels of digital rural development. The research findings reveal that arable land area, forest coverage, and mechanization play pivotal roles in shaping the trajectory of rural circular economy development. The synthetical index of rural circular economy development shows a consistent yearly increment, yet pronounced regional disparities exist, reflecting an olive-shaped distribution with dual cores in Heilongjiang and Chongqing and a more prominent central. The implementation of digital rural not only elevates the local agricultural circular economy, but also demonstrates conspicuous spatial spillover effects on neighboring regions. The relationship between digital rural development and agricultural circular economy growth is characterized by a singular threshold effect. Pre- and post-threshold, the promotive impact of digital rural development rises from 0.048 to 0.058. Building upon these findings, policy recommendations are formulated to guide future endeavors in this domain.

Crystal Polymorph Search in the NPT Ensemble via a Deposition/Sublimation Alchemical Path.

The formulation of active pharmaceutical ingredients involves discovering stable crystal packing arrangements or polymorphs, each of which has distinct pharmaceutically relevant properties. Traditional experimental screening techniques utilizing various conditions are commonly supplemented with in silico crystal structure prediction (CSP) to inform the crystallization process and mitigate risk. Predictions are often based on advanced classical force fields or quantum mechanical calculations that model the crystal potential energy landscape but do not fully incorporate temperature, pressure, or solution conditions during the search procedure. This study proposes an innovative alchemical path that utilizes an advanced polarizable atomic multipole force field to predict crystal structures based on direct sampling of the NPT ensemble. The use of alchemical (i.e., nonphysical) intermediates, a novel Monte Carlo barostat, and an orthogonal space tempering bias combine to enhance the sampling efficiency of the deposition/sublimation phase transition. The proposed algorithm was applied to 2-((4-(2-(3,4-dichlorophenyl)ethyl)phenyl)amino)benzoic acid (Cambridge Crystallography Database Centre ID: XAFPAY) as a case study to showcase the algorithm. Each experimentally determined polymorph with one molecule in the asymmetric unit was successfully reproduced via approximately 1000 short 1 ns simulations per space group where each simulation was initiated from random rigid body coordinates and unit cell parameters. Utilizing two threads of a recent Intel CPU (a Xeon Gold 6330 CPU at 2.00 GHz), 1 ns of sampling using the polarizable AMOEBA force field can be acquired in 4 h (equating to more than 300 ns/day using all 112 threads/56 cores of a dual CPU node) within the Force Field X software (https://ffx.biochem.uiowa.edu). These results demonstrate a step forward in the rigorous use of the NPT ensemble during the CSP search process and open the door to future algorithms that incorporate solution conditions using continuum solvation methods.

Accelerating atmospheric physics parameterizations using graphics processing units

As part of a project aimed at exploring the use of next-generation high-performance computing technologies for numerical weather prediction, we have ported two physics modules from the Common Community Physics Package (CCPP) to Graphics Processing Unit (GPU) and obtained accelerations of up to 10× relative to a comparable multi-core CPU. The physics parameterizations accelerated in this work are the aerosol-aware Thompson microphysics (TH) scheme and the Grell–Freitas (GF) cumulus convection scheme. Microphysics schemes are among the most time-consuming physics parameterizations, second to only radiative process schemes, and our results show better acceleration for the TH scheme than the GF scheme. Multi-GPU implementations of the schemes show acceptable weak scaling in a single node with 8 GPUs, and perfect weak scaling on multiple nodes using one GPU per node. The lack of inter-node communication for column physics parameterizations contributes to their scalability, however, physics parameterizations are run along with dynamics, so the overall multi-GPU performance is often governed by the latter. In the context of optimizing CCPP physics modules, our observations underscore that the extensive use of automatic arrays within inner subroutines hampers GPU performance due to serialized memory allocations. We have used the OpenACC directive programming language for this work because it allows for easy porting of large amounts of code and makes code maintenance more manageable compared to low-level languages like CUDA and OpenCL.

Integrated CPU Monitoring Using 2D Temperature Sensor Arrays Directly Printed on Heat Sinks

AbstractIn today's digital world, the demand for computer security and system reliability is a crucial element. Monitoring the CPU temperature during operation provides valuable insights but is currently limited to the embedded on‐chip sensors. The implementation of an extra security layer based on temperature monitoring can detect anomalies in an early stage, identify malware, and help mitigate attacks. The approach of integrating more on‐chip temperature sensors into the silicon is avoided due to space, power limitations, and cost constraints. However, the field of printed electronics and sensor technology has recently seen significant progress. This is the first work that introduces a fully‐printed temperature sensor array integrated onto the cooling unit of a CPU. It offers a novel approach for the practical implementation of such sensors into an operational computer system. The present sensor array, consisting of 396 sensor pixels, detects local hotspots induced by the underneath CPU cores, leading to an individual thermal fingerprint. This unique method paves the way for addressing pivotal elements of computer security, as deviations from expected thermal behavior can signal malicious activities. Additionally, this innovation facilitates reliability optimization. The detailed thermal map garnered provides insights into which cores are subjected to significant stress over time.

Performance Evaluation of Spintronic-Based Spiking Neural Networks using Parallel Discrete-Event Simulation

Spintronics devices that use the spin of electrons as the information state variable have the potential to emulate neuro-synaptic dynamics and can be realized within a compact form-factor, while operating at ultra-low energy-delay point. In this paper, we benchmark the performance of a spintronics hardware platform designed for handling neuromorphic tasks. To explore the benefits of spintronics-based hardware on realistic neuromorphic workloads, we developed a Parallel Discrete-Event Simulation model called Doryta, which is further integrated with a materials-to-systems benchmarking framework. The benchmarking framework allows us to obtain quantitative metrics on the throughput and energy of spintronics-based neuromorphic computing and compare these against standard CMOS-based approaches. Although spintronics hardware offers significant energy and latency advantages, we find that for larger neuromorphic circuits, the performance is limited by the interconnection networks rather than the spintronics-based neurons and synapses. This limitation can be overcome by architectural changes to the network. Through Doryta we are also able to show the power of neuromorphic computing by simulating Conway’s Game of Life (GoL), thus showing that it is Turing complete. We show that Doryta obtains over 300 × speedup using 1,024 CPU cores when tested on a convolutional, sparse, neural architecture. When scaled-up 64 times, to a 200 million neuron model, the simulation ran in 3:42 minutes for a total of 2000 virtual clock steps. The conservative approach of execution was found to be faster in most cases than the optimistic approach, even when a tie-breaking mechanism to guarantee deterministic execution, was deactivated.

Adaptive patch grid strategy for parallel protein folding using atomic burials with NAMD

The definition of protein structures is an important research topic in molecular biology currently, since there is a direct relationship between the function of the protein in the organism and the 3D geometric configuration it adopts. The transformations that occur in the protein structure from the 1D configuration to the 3D form are called protein folding. Ab initio protein folding methods use physical forces to model the interactions among the atoms that compose the protein. In order to accelerate those methods, parallel tools such as NAMD were proposed. In this paper, we propose two contributions for parallel protein folding simulations: (a) adaptive patch grid (APG) and (b) the addition of atomic burials (AB) to the traditional forces used in the simulation. With APG, we are able to adapt the simulation box (patch grid) to the current shape of the protein during the folding process. AB forces relate the 3D protein structure to its geometric center and are adequate for modeling globular proteins. Thus, adding AB to the forces used in parallel protein folding potentially increases the quality of the result for this class of proteins. APG and AB were implemented in NAMD and tested in supercomputer environments. Our results show that, with APG, we are able to reduce the execution time of the folding simulation of protein 4LNZ (5,714 atoms, 15 million time steps) from 12 hours and 36 minutes to 11 hours and 8 minutes, using 16 nodes (256 CPU cores). We also show that our APG+AB strategy was successfully used in a realistic protein folding simulation (1.7 billion time steps).