Parallel Storage Research Articles

Performance research of edge computing nodes based on heterogeneous environment

Abstract Edge computing realizes real-time parallel storage and computing in a distributed way, which can reduce the data transmission cost. The edge nodes, which are composed of different heterogeneous physical devices and hardware resources, assume the computing and storage capacity of the edge network. At present, the resource characteristics and performance of heterogeneous edge nodes are difficult to characterize. Therefore, a resource-aware performance feature detection model for heterogeneous edge compute nodes is proposed in this study. This model is based on a resource utilization matrix and resource relative performance quantization algorithm to extract the performance characteristics of the same type of resources of the edge nodes, to realize multi-dimensional and systematic feature detection and description of heterogeneous edge nodes. On this basis, a weight-based edge node task scheduling algorithm is designed. The final verification results show that the CPU relative performance ratios of Node 1-Node 2, Node 1-Node 3, and Node 2-Node 3 are 6.23, 3.81, and 0.81 respectively. The performance relationship among the three nodes is Node 1> Node 3> Node 2. The test group and the verification group tend to be consistent overall. This method has a good application prospect.

Semantic segmentation of urban land classes using a multi-scale dataset

ABSTRACT The use of remote sensing imagery for land cover and land use classification has made significant advancements in recent years. However, it becomes particularly challenging to enhance the semantic representation of high-resolution networks while dealing with uneven land categories and merging multi-scale data without compromising the accuracy of semantic segmentation. To tackle this challenge, this paper presents a novel method for classifying high-resolution remote sensing images based on a deep neural network that performs semantic segmentation of urban construction lands into five categories: vegetation, water, buildings, roads, and bare soil. The network incorporates a U-shaped high-resolution neural network and the advanced high-resolution network (HRNet) framework. The parallel storage of feature maps with different resolutions enables the exchange of information between them. The data pre-processing module addresses the issue of data imbalance in the semantic segmentation of urban construction lands, resulting in an increase in Intersection over Union (IoU) values for different land types by 3.75%-12.01%. Additionally, a target context representation module is introduced to enhance the feature representation of pixels by calculating the relationship between pixels and multiple target regions. Moreover, a polarization attention mechanism is proposed to extract the characteristics of geographical objects in all directions and achieve a stronger semantic representation. This method provides a novel approach to accurately and effectively extract information on construction lands and advance the development of monitoring algorithms for urban construction lands. To validate the proposed U-HRNet-OCR+PSA network, a comparative analysis was conducted with six classical networks, including DeepLabv3+, PSPNet, U-Net, U-Net++, HRNet, and HRNet-OCR, as well as the relatively new ViT-adapter-L, Oneformer and InternImage-H. The experiments demonstrate that the U-HRNet-OCR+PSA network achieves higher accuracy compared to the aforementioned networks. Specifically, the corresponding IoU values for the buildings, roads, vegetation, bare soil, and water in the multi-scale dataset are 89.79%, 90.05%, 94.89%, 85.91%, and 88.36%, respectively.

Retracted: Application of Internet of Things Audio Technology Based on Parallel Storage System in Music Classroom

Retracted: Application of Internet of Things Audio Technology Based on Parallel Storage System in Music Classroom

Optimum integration of latent heat storage in a solar thermal system for industrial processes: In series or in parallel?

Solar thermal energy holds great potential in decarbonizing the process industry to achieve net-zero emissions. This study focuses on analyzing the optimum integration of latent storage in solar heating and cooling for an industrial process, by comparing two configurations. Configuration 1 connects the thermal energy storage (TES) in parallel with the solar collector and the process, allowing the solar collector outlet stream to be directed to either the TES or the process. Configuration 2 connects the TES in series with the solar collector and the process, ensuring that the solar collector outlet stream always passes through the TES before reaching the process. The solar system also includes a rotary Fresnel collector that supplies heating and cooling to a dairy factory. The two configurations are evaluated by dynamic simulations including transients such as clouds and start-ups. The findings reveal that the parallel layout needs more components to achieve the same controllability and energy supply level as the series layout. Consequently, the parallel layout incurs a higher cost and complex operation. Considering the objective of finding a cost-effective and straightforward solution for the industry sector, the study concludes that connecting the TES in series represents the optimal configuration.

Magma storage and degassing beneath the youngest volcanoes of the Massif Central (France): Lessons for the monitoring of a dormant volcanic province

Developing appropriate monitoring strategies in long-quiescent volcanic provinces is challenging due to the rarity of recordable geochemical and geophysical signals and the lack of experienced eruptive phenomenology in living memory. This is the case in the Massif Central (France) where the last eruptive sequence formed the Pavin's Group of Volcanoes, about 7 ka ago. There, current evidence of a mantle activity reminiscence is suggested by the presence of mineral springwaters, mofettes, and soil degassing. It appears fundamental as a prerequisite to decipher the evolution of the gas phase in the magmatic system at the time of the eruptive activity to understand the meaning of current local gas emissions. In this study, we develop an innovative approach coupling CO2 densimetry and geochemistry of fluid inclusions from products erupted by the Pavin's Group of Volcanoes. 3D imagery by Raman spectroscopy revealed that carbonate forming in fluid inclusions may lead to underestimation of CO2 density in fluid inclusions by up to 50% and thus to unreliable barometric estimates. Fortunately, we found that this effect may be limited by focusing on fluid inclusions with a small diameter (<4 μm) and where no solid phase is detected on Raman spectra. The time evolution of the eruptions of the Pavin's Group of Volcanoes shows a progressive decrease of the pressure of magma storage (from more than 9 kbar down to 1.5–2 kbar) in parallel to magma differentiation (from basanites at Montcineyre to benmoreites at Pavin). The analysis of the noble gases entrapped in fluid inclusions yielded two main conclusions: (1) the helium isotope signature (Rc/Ra = 6.5–6.8) is in the range of values obtained in fluid inclusions from mantle xenoliths in the Massif Central (Rc/Ra = 5.6 ± 1.1, on average) suggesting partial melting of the subcontinental lithospheric mantle, and (2) magma degassing (4He/40Ar* from 4.0 to 16.2) mirrors magma differentiation and the progressive rise of the magma ponding zones of the Pavin's Group of Volcanoes. According to our modelling, about 80% of the initial gas phase would be already exsolved from these magmas, even if stored at mantle depth. Based on the results obtained from fluid inclusions, we propose a model of the evolution of the signature of noble gases and carbon isotopes from mantle depth to crustal levels. In this frame, gas emissions currently emitted in the area (Rc/Ra = 6.1–6.7 and 4He/40Ar* = 1.7) point to an origin in the lithospheric mantle. This study strongly encourages the establishment of a regular sampling of local gas emissions to detect potential geochemical variations that may reflect a change from current steady-state conditions.

Application of android voice assistant based on parallel storage in multimedia English teaching platform

Application of android voice assistant based on parallel storage in multimedia English teaching platform

Application of audio technology of Internet of Things based on parallel storage system in music classroom

Application of audio technology of Internet of Things based on parallel storage system in music classroom

Scalable Parallel Nonlinear Optimization with PyNumero and Parapint

We describe PyNumero, an open-source, object-oriented programming framework in Python that supports rapid development of performant parallel algorithms for structured nonlinear programming problems (NLP’s) using the Message Passing Interface (MPI). PyNumero provides three fundamental building blocks for developing NLP algorithms: a fast interface for calculating first and second derivatives with the AMPL Solver Library (ASL), a number of interfaces to efficient linear solvers, and block-structured vectors and matrices based on NumPy, SciPy, and MPI that support distributed parallel storage and computation. PyNumero’s design enables efficient, parallel algorithm development using high-level Python syntax while keeping expensive numerical calculations in fast, compiled implementations based on languages like C and Fortran. To demonstrate the utility of PyNumero, we also present Parapint, a Python package built on PyNumero for parallel solution of dynamic optimization problems. Parapint includes a parallel interior-point solver based on Schur-Complement decomposition. We illustrate the effectiveness of PyNumero for developing parallel algorithms with both code examples and scalability analyses for parallel matrix-vector dot products, parallel solution of structured systems of linear equations using Schur-Complement decomposition, and the parallel solution of a two-dimensional PDE optimal control problem. Our numerical results show nearly perfect scaling to more than 1,000 cores for large matrix-vector dot products and structured linear systems. Moreover, we obtain more than 354 times speedup for the optimal control example. History: Accepted by Alice Smith, EIC/Ted Ralphs, Area Editor/Software Tools. Funding: This work was funded in part by the Institute for the Design of Advanced Energy Systems (IDAES) with funding from the Office of Fossil Energy, Cross-Cutting Research, U.S. Department of Energy. This work was also funded by Sandia National Laboratories Laboratory Directed Research and Development (LDRD) program. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information ( https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2023.1272 ) as well as from the IJOC GitHub software repository ( https://github.com/INFORMSJoC/2021.0285 ) at ( http://dx.doi.org/10.5281/zenodo.7192328 ).

Research of Tri-State Optical Signal Detectors for Ternary Optical Computers

A tri-state optical signal detector (TOD) is designed in this article to improve the accuracy and stability of decoding results of the ternary optical computer (TOC). The TOD consists of three modules: a light-sensing module, a voltage divider module, and a data latching and conversion module. First, a voltage divider circuit with adjustable resistors is used to achieve parallel optical–electrical signal conversion for each LCD pixel on a photosensitive sensor plus a latch structure. Second, the problem of the parallel storage of data after photoelectric conversion is solved. Third, a transcoding circuit from the output signal of the TOD to the input signal of the tri-state optical encoder (TOE) is given. The experimental results show that the tri-state optical signal detector designed in this paper is effective and feasible.

Application of Internet of Things Audio Technology Based on Parallel Storage System in Music Classroom

In order to improve the effect of music classroom teaching, this paper combines the parallel storage system and the Internet of Things audio technology to construct a music education system to improve the effect of music education in colleges and universities. Moreover, this paper expounds a grid-free method for sparsity and parameter estimation from the model and mathematical principles and proposes a grid-free DOA method based on fast reconstruction of the T-matrix. This method is guaranteed to produce sparse parameter estimates. At the same time, the numerical simulation shows that the method has stable estimation performance as sparse and parameter estimation methods. In addition, this paper constructs a music classroom teaching system based on parallel storage system and Internet of Things audio technology. Through the experimental research, it can be seen that the parallel storage system and the Internet of Things audio technology has obvious application effect in the music classroom, which can effectively improve the teaching effect of the music classroom.

Unrepairable system with single production unit and n failure-prone identical parallel storage units

Despite the crucial role of product storage played in the operation of real-world systems, only very few reliability studies have considered this component in the system reliability modeling and analysis and the existing models assume only a single storage unit per system or subsystem. This paper makes advancements by modeling an unrepairable system with a production unit (PU) and multiple parallel storage units (SUs) of limited capacity. Both the PU and SUs are failure prone or imperfect with time-to-failure following arbitrary distributions. The surplus product in the case of the PU's performance exceeding the demand is evenly distributed among available SUs. The cumulated products in available SUs may subsequently compensate the deficiency when the PU fails or has insufficient performance. We put forward a numerical algorithm of evaluating the mission success probability (MSP) of the considered system with required demand and mission time. We conduct a case study on a water supply system with five SUs to illustrate the proposed model and investigate impacts of several model parameters (including initial SU filling, SU capacity, the number of SUs, PU productivity, reliabilities of PU and SU) as well as their interactions on the MSP.

Is the Large-Scale Development of Wind-PV with Hydro-Pumped Storage Economically Feasible in Greece?

The achievement of the long-term national energy targets in Greece for large-scale integration of wind and solar energy may be facilitated by the development of hydro-pumped storage projects. In light of the above, technical aspects related with the operation of the Greek power system and its ability to absorb renewable energy are analyzed in connection with the role of hydro-pumped storage and relative economic aspects. The aim of this work is to assess the potential contribution of hydro-pumped storage projects and estimate the capacity magnitude order to support large-scale wind and photovoltaic (PV) integration in Greece. For this purpose, scenarios for the Greek power system with focus on Wind and PV development, in conjunction with hydro-pumped storage capacity, are developed, and results for current situation and reference years 2030 and 2050 are presented. For the simulation, among others, high resolution mesoscale wind data for a typical year in the whole Greek territory are used for the steady state simulation of the Greek power system, in order to better estimate the power that could be generated from installed wind turbines, taking into consideration technical characteristics of a typical commercial wind turbine. Results indicate the need of gradual development of hydro-pumped storage in parallel with the large-scale integration of wind and PV capacity into the Greek power system. In addition, the feasibility of the examined scenarios is supported from the low cost of wind and PV generation. In the case of Greece, thanks to the complex morphology and hydraulic conditions of the country, hydro-pumped storage composes an efficient and low-cost storage solution.

Hafnium Oxide‐Based Ferroelectric Devices for Computing‐in‐Memory Applications

Herein, a layer of 10 nm ferroelectric Al‐doped HfO2 (HAO) film is fabricated and optimized and is further integrated in a nonvolatile memory device with TiN/HAO/Pt/Ti capacitor structure. Long retention, high endurance, and stable storage characteristics, as well as a competitive residual polarization of 2Pr = 24–30 μC cm−2 are achieved. Furthermore, computing‐in‐memory applications are implemented utilizing the HAO‐based ferroelectric memory devices. Typical NOR and NOT logic gates are obtained based on the memristor‐aided logic (MAGIC) operations by exploiting the polarization inversion characteristics of the device, which show great potential in realizing other basic Boolean logic operations. The results show that the HAO‐based ferroelectric memory device is a strong candidate in the pursuit of next‐generation parallel storage and computing systems.

Assessment of three solar-assisted heat pump typologies in Canada: parallel, series, and ice storage

This paper discusses the energy savings, operating costs, and net present values of three typologies of solar assisted heat pump in Canada: parallel systems, series systems, and ice storage systems. Typologies are evaluated for three detached house archetypes of varying energy performances, and across a variety of Canadian climates. Hourly energy modelling is accomplished with a custom spreadsheet tool. The models developed are approximate and meant for high level analysis. This work is meant as a first step in a process of verifying the potential for each of the typologies. In most cases, the parallel system performs best in terms of annual energy savings. The paper goes over the models, assumptions, and some results. Recommendations are discussed for future research focusing on system payback times. The next step will consist of using a detailed sub-hourly simulation tool for the typology that has been found to be the most promising.

Research on Characteristics and Programming of Simulated Human Intelligence

This paper concluded that parallel storage similar to the human brain must have the following characteristics: 1) The connection mode of its information storage is scattered and parallel. 2) The operating pattern of this system must be similar to that of the nervous system for memory, facilitation, and recall. Because of this similarity, we can “borrow” some concepts from neuroscience to describe Advanced intelligent basic programming the attributes of parallel storage systems. 3) The excitation mode of this system must be selective excitation. 4) Memory interference is one of the most important characteristics of parallel storage system. We can’t eliminate recall interference completely, but there are a number of steps can take to keep within the limits of our tolerance. 5) Memory interference requires only excitatory memory columns in the contact area. 6) The “memory excitation” mode of the parallel storage system for information must be the fuzzy “memory excitation”, which is inevitable of memory interference. Whereafter, This paper describes and discusses how and why to program on the theoretical basis of the first part. Includes an introduction to functions that enable memory, recall, stimulation, excitement. And the relationship between memory, memory, fuzzy excitement, excitement, reward and punishment, attention distribution and their significance in solving memory interference.

Review on LDPC Codes for Big Data Storage

The need for highly scalable and reliable big data storage systems is due to the fact that there is an explosive growth in data everywhere particularly due to data generated from social networking sites and IOT Technology for various applications. Hence most of Information Technology, Medical organizations and big Industries, Social networking companies, Government organizations (like ISRO) are required to have storage capacities of 100 PB (petabytes) of data. Therefore to secure and store this kind of large data efficiently research on application of Erasure codes for both Cloud storage and Network or Distributed Storage systems is considered recently. The traditional triple-replication method, which stores 3 copies of every file and requires an extra 200% storage overhead is actually highly expensive. Erasure Codes are considered for parallel storage systems as an alternative to traditional storage system. This paper presents the various techniques of applying LDPC codes for big data storage and provides the research gap to investigate application of LDPC codes for big data storage.

Understanding mechanical properties in fused filament fabrication of polyether ether ketone

Using dynamic mechanical analysis (DMA), we investigate differences in the mechanical properties of a single-filament wall of polyether ether ketone (PEEK) constructed using fused filament fabrication (FFF) under a range of different printing conditions. Since PEEK is a semi-crystalline polymer, we employ a non-isothermal quiescent crystallization model, informed by infrared (IR)-imaging measurements, to understand our findings. We propose that, under typical FFF cooling conditions, the weld region between filaments remains amorphous. In contrast, the core of the filament has increased time above the glass transition temperature allowing for a signifocant crystal fraction to develop. We correlate the predicted crystal fraction to a storage modulus using the Halpin and Kardos model. With only a single model fitting parameter we can make reasonable predictions for the perpendicular and parallel storage moduli measured via DMA over a range of printing conditions. This work provides a foundation for optimising crystallization for the mechanical performance of the FFF printed PEEK.

A popularity-aware reconstruction technique in erasure-coded storage systems

In this study, we develop a novel data reconstruction technique for parallel storage systems housed in modern data centers. We advocate for erasure-coded data storage systems to archive warm data (a.k.a., unpopular data), which attract a limited number of accesses or updates. Different from hot or cold data, warm data have to be treated in a distinctive way to optimize system performance and storage-space utilization. We pay particular attention to efficient data reconstruction in which faulty data nodes are rebuilt while responding to I/O requests. To achieve this goal, we employ two machine-learning algorithms to offer online data reconstruction in erasure coded storage systems. Our data reconstruction technique is conducive to recovering faulty nodes while boosting read performance for requests accessing data residing on the faulty nodes. Our system is reliant on a clustering mechanism to group files into multiple clusters, in each of which files share similar features. Furthermore, we implement a prediction module where a list of future popular data is projected by keeping track of historical I/O accesses. This popular-data list, in turn, provides predictions on files that are likely to be accessed in the not-too-distant future. The prediction module is responsible for computing similarities among users, thereby setting up priority levels of data blocks to be reconstructed. We implement our data reconstruction scheme in an erasure-coded parallel storage system to recover files with a guidance from the popular-data list. Our experimental results confirm that our system speeds up the data recovery of parallel storage systems while maintaining a high data access performance for on-line users.

THE CONTRIBUTION OF SHORT-TERM MEMORY, WORKING MEMORY AND DEMOGRAPHIC CHARACTERISTICS TO THE LINGUISTIC ACHIEVEMENT OF GREEK MIDDLE SCHOOL-AGE CHILDREN: IMPLICATIONS FOR NATIVE LANGUAGE DEVELOPMENT

The present study delved into the involvement of Greek-speaking middle primary school children’s mnemonic recourses and demographic features in their performance in language tests. One hundred and seventy-six students aged 8.50-12.05 (mean age = 10.11 years) participated. The native language of the sample differed from English, in which most investigations have been conducted. Children’s verbal short-term memory (STM) and working memory (WM) capacity were assessed with a simple verbal retention measure (straight digit recall) and a task requiring parallel storage and processing of verbal information (backward digit recall), respectively. Pupils’ vocabulary knowledge was estimated through an assignment of expressive vocabulary. Children were also administered a reading fluency and a reading comprehension task. Pearson’s r and Spearman’s rho correlation coefficients showed that students’ verbal STM and WM significantly correlated with their reading fluency and text comprehension, accordingly. In addition, by employing simple linear regression analysis, increase in participants’ age was found to predict their extended reading fluency, while one-way Analysis of Variance (ANOVA) of independent samples revealed that vocabulary knowledge varied, being subject to pupils’ school grade. Lastly, independent samples t- test demonstrated no substantial difference in reading fluency between the two genders of children. Current findings are discussed in relation to participants’ age growth, their cumulating linguistic experience and the historical properties of the Greek language, indicating the need for further research that takes into account the unique and multifaceted nature of readers’ profile and outlining a few connotations for literacy instruction. Article visualizations:

Data Optimization using Apache Flink

Map Reduce, Flink, and Spark, also become more popular in the processing of big data lately. Flink will be an open platform Big Data processing system for Apache-powered batch storage and streaming of data. Flink's query optimizer is constructed for historical information processing (batch) based on parallel storage systems approaches. Flink query query optimizer interprets the questions into jobs of different tasks that are regularly sent. Therefore, taking advantage of task similarities should prevent redundant computation. In this article, the multi-demand optimization model for Flink, Flink was planned and designed on Flink Software Stack's top priority. It's thought-about as an associate in Apache Flink's nursing add-on to maximize multi-demand information sharing. The Flink system takes advantage of option operators ' information sharing resources to reduce overlap and duplication of multi-query in-network information movement. Research findings show that the leveraging of shared option operations in vast information on multiple requests would offer promising time to perform queries. Therefore, in the stream phase, Without doubt the Flink approach can be used to boost application performance over time periods.