Parallel Computational Resources Research Articles

Application of an Improved Workflow for the Identification of Flame Dynamics to Swirl Stabilized Wet Combustion

Abstract The estimation of flame transfer functions (FTF) from time series data generated by large eddy simulation (LES) via system identification (SI) is an important element of thermoacoustic analysis. A continuous time series of adequate length is required to achieve low uncertainty, especially when dealing with turbulent noise. Limited scalability of LES codes implies that the wall-clock-time required for generating such time series may be excessive. The present paper tackles this challenge by exploring how the superposition of multiple simulations with the same excitation signal, but varying initial conditions, increases signal-to-noise ratio and leads to more robust identification. In addition, the established SI approach, which relies on broadband excitation, is compared to excitation with approximate Dirac and Heaviside signals, promising simpler pre- and postprocessing. Results demonstrate that the proposed workflow reduces significantly the wall-clock-time required for robust FTF identification. This reduction in wall-clock-time requires more parallel computational resources, but it does not significantly increase the overall computational cost while also enabling FTF estimation using Heaviside excitation. The proposed method is assessed on a partially premixed, steam enriched (‘WET’) swirl burner with significant turbulent noise levels. Steam enrichment is a combustion concept that reduces harmful emissions such as NOx and CO2 while increasing engine efficiency. However, the effect of steam on the flame response, needs to be better understood. To this end, a combustion model including an optimized global chemical mechanism for partially premixed wet methane combustion, is presented and validated against experimental data.

BDIS-SLAM: a lightweight CPU-based dense stereo SLAM for surgery.

Common dense stereo simultaneous localization and mapping (SLAM) approaches in minimally invasive surgery (MIS) require high-end parallel computational resources for real-time implementation. Yet, it is not always feasible since the computational resources should be allocated to other tasks like segmentation, detection, and tracking. To solve the problem of limited parallel computational power, this research aims at a lightweight dense stereo SLAM system that works on a single-core CPU and achieves real-time performance (more than 30Hz in typical scenarios). A new dense stereo mapping module is integrated with the ORB-SLAM2 system and named BDIS-SLAM. Our new dense stereo mapping module includes stereo matching and 3D dense depth mosaic methods. Stereo matching is achieved with the recently proposed CPU-level real-time matching algorithm Bayesian Dense Inverse Searching (BDIS). A BDIS-based shape recovery and a depth mosaic strategy are integrated as a new thread and coupled with the backbone ORB-SLAM2 system for real-time stereo shape recovery. Experiments on in vivo data sets show that BDIS-SLAM runs at over 30Hz speed on modern single-core CPU in typical endoscopy/colonoscopy scenarios. BDIS-SLAM only consumes around an additional time compared with the backbone ORB-SLAM2. Although our lightweight BDIS-SLAM simplifies the process by ignoring deformation and fusion procedures, it can provide a usable dense mapping for modern MIS on computationally constrained devices. The proposed BDIS-SLAM is a lightweight stereo dense SLAM system for MIS. It achieves 30Hz on a modern single-core CPU in typical endoscopy/colonoscopy scenarios (image size around ). BDIS-SLAM provides a low-cost solution for dense mapping in MIS and has the potential to be applied in surgical robots and AR systems. Code is available at https://github.com/JingweiSong/BDIS-SLAM .

Resource-Aware Distributed Differential Evolution for Training Expensive Neural-Network-Based Controller in Power Electronic Circuit.

The neural-network (NN)-based control method is a new emerging promising technique for controller design in a power electronic circuit (PEC). However, the optimization of NN-based controllers (NNCs) has significant challenges in two aspects. The first challenge is that the search space of the NNC optimization problem is such complex that the global optimization ability of the existing algorithms still needs to be improved. The second challenge is that the training process of the NNC parameters is very computationally expensive and requires a long execution time. Thus, in this article, we develop a powerful evolutionary computation-based algorithm to find a high-quality solution and reduce computational time. First, the differential evolution (DE) algorithm is adopted because it is a powerful global optimizer in solving a complex optimization problem. This can help to overcome the premature convergence in local optima to train the NNC parameters well. Second, to reduce the computational time, the DE is extended to distribute DE (DDE) by dispatching all the individuals to different distributed computing resources for parallel computing. Moreover, a resource-aware strategy (RAS) is designed to further efficiently utilize the resources by adaptively dispatching individuals to resources according to the real-time performance of the resources, which can simultaneously concern the computing ability and load state of each resource. Experimental results show that, compared with some other typical evolutionary algorithms, the proposed algorithm can get significantly better solutions within a shorter computational time.

An aerosol vertical data assimilation system (NAQPMS-PDAF v1.0): development and application

Abstract. Aerosol vertical stratification is important for global climate and planetary boundary layer (PBL) stability, and no single method can obtain spatiotemporally continuous vertical profiles. This paper develops an online data assimilation (DA) framework for the Eulerian atmospheric chemistry-transport model (CTM) Nested Air Quality Prediction Model System (NAQPMS) with the Parallel Data Assimilation Framework (PDAF) as the NAQPMS-PDAF for the first time. Online coupling occurs based on a memory-based way with two-level parallelization, and the arrangement of state vectors during the filter is specifically designed. Scaling tests demonstrate that the NAQPMS-PDAF can make efficient use of parallel computational resources for up to 25 000 processors with a weak scaling efficiency of up to 0.7. The 1-month long aerosol extinction coefficient profiles measured by the ground-based lidar and the concurrent hourly surface PM2.5 are solely and simultaneously assimilated to investigate the performance and application of the DA system. The hourly analysis and subsequent 1 h simulation are validated through lidar and surface PM2.5 measurements assimilated and not assimilated. The results show that lidar DA can significantly improve the underestimation of aerosol loading, especially at a height of approximately 400 m in the free-running (FR) experiment, with the mean bias (BIAS) changing from −0.20 (−0.14) km−1 to −0.02 (−0.01) km−1 and correlation coefficients increasing from 0.33 (0.28) to 0.91 (0.53) averaged over sites with measurements assimilated (not assimilated). Compared with the FR experiment, simultaneously assimilating PM2.5 and lidar can have a more consistent pattern of aerosol vertical profiles with a combination of surface PM2.5 and lidar, independent extinction coefficients from the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP), and aerosol optical depth (AOD) from the Aerosol Robotic Network (AERONET). Lidar DA has a larger temporal impact than that in PM2.5 DA but has deficiencies in subsequent quantification on the surface PM2.5. The proposed NAQPMS-PDAF has great potential for further research on the impact of aerosol vertical distribution.

ExpertRNA: A New Framework for RNA Secondary Structure Prediction

Ribonucleic acid (RNA) is a fundamental biological molecule that is essential to all living organisms, performing a versatile array of cellular tasks. The function of many RNA molecules is strongly related to the structure it adopts. As a result, great effort is being dedicated to the design of efficient algorithms that solve the “folding problem”—given a sequence of nucleotides, return a probable list of base pairs, referred to as the secondary structure prediction. Early algorithms largely rely on finding the structure with minimum free energy. However, the predictions rely on effective simplified free energy models that may not correctly identify the correct structure as the one with the lowest free energy. In light of this, new, data-driven approaches that not only consider free energy, but also use machine learning techniques to learn motifs are also investigated and recently been shown to outperform free energy–based algorithms on several experimental data sets. In this work, we introduce the new ExpertRNA algorithm that provides a modular framework that can easily incorporate an arbitrary number of rewards (free energy or nonparametric/data driven) and secondary structure prediction algorithms. We argue that this capability of ExpertRNA has the potential to balance out different strengths and weaknesses of state-of-the-art folding tools. We test ExpertRNA on several RNA sequence-structure data sets, and we compare the performance of ExpertRNA against a state-of-the-art folding algorithm. We find that ExpertRNA produces, on average, more accurate predictions of nonpseudoknotted secondary structures than the structure prediction algorithm used, thus validating the promise of the approach. Summary of Contribution: ExpertRNA is a new algorithm inspired by a biological problem. It is applied to solve the problem of secondary structure prediction for RNA molecules given an input sequence. The computational contribution is given by the design of a multibranch, multiexpert rollout algorithm that enables the use of several state-of-the-art approaches as base heuristics and allowing several experts to evaluate partial candidate solutions generated, thus avoiding assuming the reward being optimized by an RNA molecule when folding. Our implementation allows for the effective use of parallel computational resources as well as to control the size of the rollout tree as the algorithm progresses. The problem of RNA secondary structure prediction is of primary importance within the biology field because the molecule structure is strongly related to its functionality. Whereas the contribution of the paper is in the algorithm, the importance of the application makes ExpertRNA a showcase of the relevance of computationally efficient algorithms in supporting scientific discovery.

Distributed Real-Time Image Processing of Formation Flying SAR Based on Embedded GPUs

<b>Formation flying synthetic aperture radar (FF-SAR) systems, as an important development direction of multichannel SAR, can achieve high-resolution wide-swath (HRWS) imaging. Coherently combining data from satellite receivers puts a strain on traditional real-time processing systems based on individual satellites. Characteristics such as the power of real-time on-orbit processing platform must be properly balanced with constrained memory and parallel computational resources. This paper proposes a distributed SAR real-time imaging method based on embedded Graphics Processing Units (GPUs). The parallel computing method of chirp scaling (CS) algorithm is designed based on parallel programming model of compute unified device architecture (CUDA), and the optimization methods of memory and performance are proposed for the hardware architecture of embedded GPUs. In particular, unified memory management method is used to avoid data copying and communication delays between the CPU and GPU. A hardware verification system for distributed SAR real-time imaging processing based on multiple embedded GPUs is constructed. The proposed algorithm takes 5.86 seconds to process single-precision floating-point complex imaging with a data size of 8192 &#x00D7; 8192 on a single Jetson Nano platform. The actual power consumption is less than 5 W, and the performance-to-power ratio is greater than 1.7&#x0025;. The experimental results show that the real-time processing method based on embedded GPUs proposed in this paper has high performance and low power consumption.</b>

Widening: using parallel resources to improve model quality

This paper provides a unified description of Widening, a framework for the use of parallel (or otherwise abundant) computational resources to improve model quality. We discuss different theoretical approaches to Widening with and without consideration of diversity. We then soften some of the underlying constraints so that Widening can be implemented in real world algorithms. We summarize earlier experimental results demonstrating the potential impact as well as promising implementation strategies before concluding with a survey of related work.

Hyperspectral Parallel Image Compression on Edge GPUs

Edge applications evolved into a variety of scenarios that include the acquisition and compression of immense amounts of images acquired in space remote environments such as satellites and drones, where characteristics such as power have to be properly balanced with constrained memory and parallel computational resources. The CCSDS-123 is a standard for lossless compression of multispectral and hyperspectral images used in on-board satellites and military drones. This work explores the performance and power of 3 families of low-power heterogeneous Nvidia GPU Jetson architectures, namely the 128-core Nano, the 256-core TX2 and the 512-core Xavier AGX by proposing a parallel solution to the CCSDS-123 compressor on embedded systems, reducing development effort, compared to the production of dedicated circuits, while maintaining low power. This solution parallelizes the predictor on the low-power GPU while the entropy encoders exploit the heterogeneous multiple CPU cores and the GPU concurrently. We report more than 4.4 GSamples/s for the predictor and up to 6.7 Gb/s for the complete system, requiring less than 11 W and providing an efficiency of 611 Mb/s/W.

High-efficiency simulation of VENUS-3 neutron-shielding problem with an automatic and enhanced hybrid Monte-Carlo-Deterministic method

To solve the deep-penetration problem in the Monte-Carlo simulation, the classic consistent adjoint driven importance sampling (CADIS) and forward-weighted CADIS (FW-CADIS) hybrid MC-deterministic method is adopted and enhanced in this paper. An automatic hybrid MC-deterministic code, NECP-MCX, is newly developed. NECP-MCX embeds the discrete-ordinate code NECP-Hydra to automatically perform the forward and adjoint neutron-transport calculations and set up the importance parameters for the MC simulation. In massive parallel calculation for large shielding problems, huge amounts of memory are required to save detailed weight-window information. To mitigate the memory limitation, a mesh-coarsening algorithm is developed based on the contributon theory. It can be applied in the traditional Cartesian-mesh and the nested-mesh structure. NECP-MCX was applied to the VENUS-3 benchmark. The results show that NECP-MCX with the enhanced hybrid method is effective in variance reduction and memory saving for the simulation of large and complicated shielding problems on parallel computer resources.

An asynchronous distributed optimization method for energy saving of parallel-connected pumps in HVAC systems

The energy consumption of parallel-connected pumps in heating, ventilation, and air-conditioning (HVAC) systems make up a large proportion of building energy consumption leading to an increasing attention for the pump operation. The centralized and distributed methods have been proposed to solve the pump optimization problem, which need the global clock to be synchronized and lead to the increasing system cost and idle time. In this paper, we propose an asynchronous distributed control algorithm for parallel-connected pumps in HVAC systems, which does not need the global clock, reduces the idle time and has the convergence guarantee. Under the bounded delay, the algorithm makes the optimal control decision for pumps to minimize energy consumption and meet the system demand. The theoretical analysis on convergence of the algorithm is established. The whole system is fully distributed and each pump is controlled by an intelligent node that runs identical control code and does not need to be synchronous. Numerical simulations on 6 parallel-connected pumps for different working cases are given to illustrate the effectiveness of the proposed algorithm and compare with the method from other field. The results show that our method strictly meets the constraint and minimizes system energy consumption, fully uses parallel computational resources. The running time of our asynchronous method is near 46% faster than synchronous setting.

Gaussbock: Fast Parallel-iterative Cosmological Parameter Estimation with Bayesian Nonparametrics

Abstract We present and apply Gaussbock, a new embarrassingly parallel iterative algorithm for cosmological parameter estimation designed for an era of cheap parallel-computing resources. Gaussbock uses Bayesian nonparametrics and truncated importance sampling to accurately draw samples from posterior distributions with an orders-of-magnitude speed-up in wall time over alternative methods. Contemporary problems in this area often suffer from both increased computational costs due to high-dimensional parameter spaces and consequent excessive time requirements, as well as the need to fine-tune proposal distributions or sampling parameters. Gaussbock is designed specifically with these issues in mind. We explore and validate the performance and convergence of the algorithm on a fast approximation to the Dark Energy Survey Year 1 (DES Y1) posterior, finding reasonable scaling behavior with the number of parameters. We then test on the full DES Y1 posterior using large-scale supercomputing facilities and recover reasonable agreement with previous chains, although the algorithm can underestimate the tails of poorly constrained parameters. Additionally, we discuss and demonstrate how Gaussbock recovers complex posterior shapes very well at lower dimensions, but faces challenges to perform well on such distributions in higher dimensions. In addition, we provide the community with a user-friendly software tool for accelerated cosmological parameter estimation based on the methodology described in this paper.

Adaptive Inattentional Framework for Video Object Detection With Reward-Conditional Training

Recent object detection studies have been focused on video sequences, mostly due to the increasing demand of industrial applications. Although single-image architectures achieve remarkable results in terms of accuracy, they do not take advantage of particular properties of the video sequences and usually require high parallel computational resources, such as desktop GPUs. In this work, an inattentional framework is proposed, where the object context in video frames is dynamically reused in order to reduce the computation overhead. The context features corresponding to keyframes are fused into a synthetic feature map, which is further refined using temporal aggregation with ConvLSTMs. Furthermore, an inattentional policy has been learned to adaptively balance the accuracy and the amount of context reused. The inattentional policy has been learned under the reinforcement learning paradigm, and using our novel reward-conditional training scheme, which allows for policy training over a whole distribution of reward functions and enables the selection of a unique reward function at inference time. Our framework shows outstanding results on platforms with reduced parallelization capabilities, such as CPUs, achieving an average latency reduction up to 2.09×, and obtaining FPS rates similar to their equivalent GPU platform, at the cost of a 1.11× mAP reduction.

PaCBAM: fast and scalable processing of whole exome and targeted sequencing data

BackgroundInterrogation of whole exome and targeted sequencing NGS data is rapidly becoming a preferred approach for the exploration of large cohorts in the research setting and importantly in the context of precision medicine. Single-base and genomic region level data retrieval and processing still constitute major bottlenecks in NGS data analysis. Fast and scalable tools are hence needed.ResultsPaCBAM is a command line tool written in C and designed for the characterization of genomic regions and single nucleotide positions from whole exome and targeted sequencing data. PaCBAM computes depth of coverage and allele-specific pileup statistics, implements a fast and scalable multi-core computational engine, introduces an innovative and efficient on-the-fly read duplicates filtering strategy and provides comprehensive text output files and visual reports. We demonstrate that PaCBAM exploits parallel computation resources better than existing tools, resulting in important reductions of processing time and memory usage, hence enabling an efficient and fast exploration of large datasets.ConclusionsPaCBAM is a fast and scalable tool designed to process genomic regions from NGS data files and generate coverage and pileup comprehensive statistics for downstream analysis. The tool can be easily integrated in NGS processing pipelines and is available from Bitbucket and Docker/Singularity hubs.

Data-Driven Network Intelligence for Anomaly Detection

Data-driven network intelligence will offer a robust, efficient, and effective computing system for anomaly detection in cyber security applications. In this article, we first summarize the current development and challenges of network intelligence for anomaly detection. Based on the current development, we propose a data-driven intelligence system for network anomaly detection. With the support of extended computing, storage, and other resources to the network edge, fog computing is incorporated into the design of the system. The proposed system consists of three major components: fog enabled infrastructure, fog enabled artificial intelligence (AI) engine, and threat intelligence. Fog enabled infrastructure provides efficient and effective computing resources for parallel computing and data storage. The fog enabled AI engine produces optimal learning models for threat detection, and enables efficient model update both locally and globally. Threat intelligence offers real-time network monitoring and cyber threat detection. We demonstrate that the proposed data-driven network intelligence system achieves high detection accuracy and provides efficient computational performance.

On the use of many-core machines for the acceleration of a mesh truncation technique for FEM

Finite element method (FEM) has been used for years for radiation problems in the field of electromagnetism. To tackle problems of this kind, mesh truncation techniques are required, which may lead to the use of high computational resources. In fact, electrically large radiation problems can only be tackled using massively parallel computational resources. Different types of multi-core machines are commonly employed in diverse fields of science for accelerating a number of applications. However, properly managing their computational resources becomes a very challenging task. On the one hand, we present a hybrid message passing interface + OpenMP-based acceleration of a mesh truncation technique included in a FEM code for electromagnetism in a high-performance computing cluster equipped with 140 compute nodes. Results show that we obtain about 85% of the theoretical maximum speedup of the machine. On the other hand, a graphics processing unit has been used to accelerate one of the parts that presents high fine-grain parallelism.

Allele-Specific Expression Analysis in Cancer Using Next-Generation Sequencing Data.

Allele-specific expression arises when transcriptional activity at the different alleles of a gene differs considerably. Although extensive research has been carried out to detect and characterize this phenomenon, the landscape of allele-specific expression in cancer is still poorly understood. In this chapter, we describe a fast and reliable analysis pipeline to study allele-specific expression in cancer using next-generation sequencing data. The pipeline provides a gene-level analysis approach that exploits paired germline DNA and tumor RNA sequencing data and benefits from parallel computation resources when available.

Massively parallel multicanonical simulations

Generalized-ensemble Monte Carlo simulations such as the multicanonical method and similar techniques are among the most efficient approaches for simulations of systems undergoing discontinuous phase transitions or with rugged free-energy landscapes. As Markov chain methods, they are inherently serial computationally. It was demonstrated recently, however, that a combination of independent simulations that communicate weight updates at variable intervals allows for the efficient utilization of parallel computational resources for multicanonical simulations. Implementing this approach for the many-thread architecture provided by current generations of graphics processing units (GPUs), we show how it can be efficiently employed with of the order of 104 parallel walkers and beyond, thus constituting a versatile tool for Monte Carlo simulations in the era of massively parallel computing. We provide the fully documented source code for the approach applied to the paradigmatic example of the two-dimensional Ising model as starting point and reference for practitioners in the field. Program summaryProgram Title: cudamucaProgram Files doi:http://dx.doi.org/10.17632/tzhfpdymv9.1Licensing provisions: Creative Commons Attribution license (CC BY 4.0)Programming language: C++, CUDAExternal routines/libraries: NVIDIA CUDA Toolkit 6.5 or newerNature of problem: The program determines weights for a multicanonical simulation of the 2D Ising model to result in a flat energy histogram. A final production run with these weights provides an estimate of the density of states of the model.Solution method: The code uses a parallel variant of the multicanonical method employing many parallel walkers that accumulate a common histogram. The resulting histogram is used to determine the weight function for the next iteration. Once the iteration has converged, simulations visit all possible energies with the same probability.Additional comments including restrictions and unusual features: The system size and size of the population of replicas are limited depending on the memory of the GPU device used. Code repository at https://github.com/CQT-Leipzig/cudamuca.

A multipurpose computing center with distributed resources

The Computing Center of the Institute of Physics (CC IoP) of the Czech Academy of Sciences serves a broad spectrum of users with various computing needs. It runs WLCG Tier-2 center for the ALICE and the ATLAS experiments; the same group of services is used by astroparticle physics projects the Pierre Auger Observatory (PAO) and the Cherenkov Telescope Array (CTA). OSG stack is installed for the NOvA experiment. Other groups of users use directly local batch system. Storage capacity is distributed to several locations. DPM servers used by the ATLAS and the PAO are all in the same server room, but several xrootd servers for the ALICE experiment are operated in the Nuclear Physics Institute in Řež, about 10 km away. The storage capacity for the ATLAS and the PAO is extended by resources of the CESNET - the Czech National Grid Initiative representative. Those resources are in Plzen and Jihlava, more than 100 km away from the CC IoP. Both distant sites use a hierarchical storage solution based on disks and tapes. They installed one common dCache instance, which is published in the CC IoP BDII. ATLAS users can use these resources using the standard ATLAS tools in the same way as the local storage without noticing this geographical distribution. Computing clusters LUNA and EXMAG dedicated to users mostly from the Solid State Physics departments offer resources for parallel computing. They are part of the Czech NGI infrastructure MetaCentrum with distributed batch system based on torque with a custom scheduler. Clusters are installed remotely by the MetaCentrum team and a local contact helps only when needed. Users from IoP have exclusive access only to a part of these two clusters and take advantage of higher priorities on the rest (1500 cores in total), which can also be used by any user of the MetaCentrum. IoP researchers can also use distant resources located in several towns of the Czech Republic with a capacity of more than 12000 cores in total.

A methodology for reduced order modeling and calibration of the upper atmosphere

AbstractAtmospheric drag is the largest source of uncertainty in accurately predicting the orbit of satellites in low Earth orbit (LEO). Accurately predicting drag for objects that traverse LEO is critical to space situational awareness. Atmospheric models used for orbital drag calculations can be characterized either as empirical or physics‐based (first principles based). Empirical models are fast to evaluate but offer limited real‐time predictive/forecasting ability, while physics based models offer greater predictive/forecasting ability but require dedicated parallel computational resources. Also, calibration with accurate data is required for either type of models. This paper presents a new methodology based on proper orthogonal decomposition toward development of a quasi‐physical, predictive, reduced order model that combines the speed of empirical and the predictive/forecasting capabilities of physics‐based models. The methodology is developed to reduce the high dimensionality of physics‐based models while maintaining its capabilities. We develop the methodology using the Naval Research Lab's Mass Spectrometer Incoherent Scatter model and show that the diurnal and seasonal variations can be captured using a small number of modes and parameters. We also present calibration of the reduced order model using the CHAMP and GRACE accelerometer‐derived densities. Results show that the method performs well for modeling and calibration of the upper atmosphere.

Unique cavity-based operator and hierarchical domain partitioning for fast parallel generation of anisotropic meshes

We devise a strategy in order to generate large-size adapted tetrahedral anisotropic meshes, having O(108–109) elements, as required in many fields of application in scientific computing. We target moderate scale parallel computational resources as typically found in R&D units where the number of cores ranges in 102–103. Both distributed and shared memory architectures are handled. Our strategy is based on typical domain splitting algorithm where the initial mesh is split into parts that are then meshed in parallel while the fictitious boundaries between parts are kept unchanged. Then we iterate the procedure to adapt previously unmodified parts of the domain, i.e., the interface mesh. Both the volume and the surface meshes are adapted simultaneously and the efficiency of the method is independent of the complexity of the geometry. The originality of the method relies on (i) a metric-based static load-balancing, (ii) hierarchical mesh partitioning techniques to (re)split the (complex) interfaces meshes, (iii) a fast, robust and generic sequential cavity-based mesh modification kernel. In order to generate large-size meshes, out-of-core storing of completed parts is used to reduce the memory footprint. We show that we are able to generate (uniform, isotropic and anisotropic) meshes with more than 1 billion tetrahedra in less than 20 minutes on 120 cores. Examples from Computational Fluid Dynamics (CFD) simulations are also discussed.