Reduce Data Communication Research Articles

Signal Processing Method to Reduce Data Communication Delay between Computer and PLC

Signal Processing Method to Reduce Data Communication Delay between Computer and PLC

Clustering-based multi-objective optimization considering fairness for multi-workflow scheduling on clouds

Distributed computing, such as cloud computing, provides promising platforms for orchestrating scientific workflows' tasks based on their sequences and dependencies. Workflow scheduling plays an important role in optimizing concerned objectives for distributed computing, such as minimizing the makespan and cost. Many researchers have focused on optimizing a specific single workflow with multiple objectives. Currently, there are few studies on multi-workflow scheduling, with most research focusing on objectives such as cost and makespan. However, multi-workflow scheduling requires the design of specific objectives that reflect the unique characteristics of multiple workflows. On the other hand, clustering-based approaches have garnered significant attention in the field of workflow scheduling over distributed computing resources due to their advantage in reducing data communication among tasks. Despite this, the effectiveness of clustering-based algorithms has not been extensively studied and validated in the context of multi-objective multi-workflow scheduling models. Motivated by these factors, we propose an approach for multiple workflows' multi-objective optimization (MOO), considering the new defined metric, fairness. We first mathematically formulate the fairness and define a fairness-involved MOO model. Then, we propose an advanced clustering-based resource optimization strategy in multiple workflow runs. Experimental results show that the proposed approach performs better than the compared algorithms without significant compromise of the overall makespan and cost as well as individual fairness, which can guide the simulation workflow scheduling on clouds.

Particle-resolved thermal lattice Boltzmann simulation using OpenACC on multi-GPUs

We utilize the Open Accelerator (OpenACC) approach for graphics processing unit (GPU) accelerated particle-resolved thermal lattice Boltzmann (LB) simulation. We adopt the momentum-exchange method to calculate fluid-particle interactions to preserve the simplicity of the LB method. To address load imbalance issues, we extend the indirect addressing method to collect fluid-particle link information at each timestep and store indices of fluid-particle link in a fixed index array. We simulate the sedimentation of 4,800 hot particles in cold fluids with a domain size of 40002, and the simulation achieves 1750 million lattice updates per second (MLUPS) on a single GPU. Furthermore, we implement a hybrid OpenACC and message passing interface (MPI) approach for multi-GPU accelerated simulation. This approach incorporates four optimization strategies, including building domain lists, utilizing request-answer communication, overlapping communications with computations, and executing computation tasks concurrently. By reducing data communication between GPUs, hiding communication latency through overlapping computation, and increasing the utilization of GPU resources, we achieve improved performance, reaching 10846 MLUPS using 8 GPUs. Our results demonstrate that the OpenACC-based GPU acceleration is promising for particle-resolved thermal lattice Boltzmann simulation.

GraphAGILE: An FPGA-Based Overlay Accelerator for Low-Latency GNN Inference

This paper presents GraphAGILE, a domain-specific FPGA-based overlay accelerator for graph neural network (GNN) inference. GraphAGILE consists of (1) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a novel unified architecture design</i> with an <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">instruction set</i> , and (2) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a compiler</i> built upon the instruction set that can quickly generate optimized code. Due to the proposed instruction set architecture (ISA) and the compiler, GraphAGILE does not require any FPGA reconfiguration when performing inference on various GNN models and input graphs. For the architecture design, we propose a novel hardware module named Adaptive Computation Kernel (ACK), that can execute various computation kernels of GNNs, including general matrix multiplication (GEMM), sparse-dense matrix multiplication (SpDMM), and sampled dense-dense matrix multiplication (SDDMM). The compiler takes the specifications of a GNN model and the graph meta data (e.g., the number of vertices and edges) as input, and generates a sequence of instructions for inference execution. We develop the following compiler optimizations to reduce inference latency: (1) computation order optimization that automatically reorders the computation graph to reduce the total computation complexity, (2) layer fusion that merges adjacent layers to reduce data communication volume, (3) data partitioning with a partition-centric execution scheme that partitions the input graph to fit the available on-chip memory of FPGA, (4) kernel mapping that automatically selects execution mode for ACK, and performs task scheduling to overlap computation with data communication and achieves dynamic load balance. We implement GraphAGILE on a state-of-the-art FPGA platform, Xilinx Alveo U250. GraphAGILE can execute widely used GNN models, including GCN, GAT, GIN, GraphSAGE, SGC and other GNN models supported by GraphGym. Experimental results show that GraphAGILE achieves up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$47.1\times$</tex-math></inline-formula> ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$3.9\times$</tex-math></inline-formula> ) reduction in end-to-end latency, including the latency of compilation and hardware execution, compared with the state-of-the-art implementations on CPU (GPU), and achieves up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$2.9\times$</tex-math></inline-formula> reduction in hardware execution latency compared with the state-of-the-art FPGA accelerators.

Towards an Effective Service Allocation in Fog Computing.

The Internet of Things (IoT) generates a large volume of data whenever devices are interconnected and exchange data across a network. Consequently, a variety of services with diverse needs arises, including capacity requirements, data quality, and latency demands. These services operate on fog computing devices, which are limited in power and bandwidth compared to the cloud. The primary challenge lies in determining the optimal location for service implementation: in the fog, in the cloud, or in a hybrid setup. This paper introduces an efficient allocation technique that moves processing closer to the network's fog side. It explores the optimal allocation of devices and services while maintaining resource utilization within an IoT architecture. The paper also examines the significance of allocating services to devices and optimizing resource utilization in fog computing. In IoT scenarios, where a wide range of services and devices coexist, it becomes crucial to effectively assign services to devices. We propose priority-based service allocation (PSA) and sort-based service allocation (SSA) techniques, which are employed to determine the optimal order for the utilizing devices to perform different services. Experimental results demonstrate that our proposed technique reduces data communication over the network by 88%, which is achieved by allocating most services locally in the fog. We increased the distribution of services to fog devices by 96%, while simultaneously minimizing the wastage of fog resources.

A Fast Parallel Random Forest Algorithm Based on Spark

To improve the computational efficiency and classification accuracy in the context of big data, an optimized parallel random forest algorithm is proposed based on the Spark computing framework. First, a new Gini coefficient is defined to reduce the impact of feature redundancy for higher classification accuracy. Next, to reduce the number of candidate split points and Gini coefficient calculations for continuous features, an approximate equal-frequency binning method is proposed to determine the optimal split points efficiently. Finally, based on Apache Spark computing framework, the forest sampling index (FSI) table is defined to speed up the parallel training process of decision trees and reduce data communication overhead. Experimental results show that the proposed algorithm improves the efficiency of constructing random forests while ensuring classification accuracy, and is superior to Spark-MLRF in terms of performance and scalability.

Distributed data filtering and modeling for fog and networked manufacturing

Fog Manufacturing applies both Fog and Cloud Computing collaboratively in Smart Manufacturing to create an interconnected network through sensing, actuation, and computation nodes. Fog Manufacturing has become a promising research component to be integrated into the existing Smart Manufacturing paradigm and provides reliable and responsive computation services. However, Fog nodes' relatively limited communication bandwidth and computation capabilities call for reduced data communication load and computation time latency for modeling. There has long been a lack of an integrated framework to automatically reduce manufacturing data and perform computationally efficient modeling/machine learning. This research direction is increasingly important as both the computational demands and Fog/networked Manufacturing become prevalent. This paper proposes an integrated and distributed framework for data reduction and modeling of multiple systems in a Smart Manufacturing network considering the system similarities. A simulation study and a Fog Manufacturing testbed for ingot growth manufacturing validated that the proposed framework significantly reduces the sample size used for improved computational runtime metrics while outperforming various other data reduction methods in modeling performance.

Fed-IoUT: Opportunities and Challenges of Federated Learning in the Internet of Underwater Things

The Internet of Underwater Things (IoUT), as an emerging technology, plays an important role in the development of marine resources. With the growth of IoUT equipment, data from various sources has increased dramatically. Multiple challenges, such as data privacy, communication overhead, and data heterogeneity, arise with the application of new underwater Internet of Things (IoT) scenarios. To solve these challenges, we propose an IoUT framework called Fed-IoUT, which utilizes Federated Learning (FL) paradigm. FL, with its advantages of data privacy protection, low communication overhead, and balanced computing resources, is perfectly suited to our scenario. Fed-IoUT can make full use of the large amount of data collected by IoUT equipment, reduce data communication overhead, and protect the privacy of observing data. In this article, we also introduced the application of Fed-IoUT in relevant fields from five aspects: water quality monitoring, underwater life detection, flood forecasting, underwater detection, and marine energy transfer. And then, we put forward the limitations of development under the current conditions and made a comprehensive outlook for the future.

Sustainable and Efficient Fog-Assisted IoT Cloud Based Data Collection and Delivery for Smart Cities

The Internet of Things (IoT) aims to enhance the quality of human life by sensing, processing and delivering data from the physical world. The Internet of Things Cloud (IoTC) is able to enhance sensing capabilities of IoT devices while fog computing can overcome computation and storage limitations of IoT devices. Based on this observation, we propose a fog-assisted IoTC data collection and delivery framework to reduce IoTC-based data communication costs and enhance success rates. The framework exploits the named data networking to deliver IoTC-based data so that multiple IoT devices can retrieve data from the nearest provider via one data delivery process. Moreover, the framework achieves mobility support for IoT devices to guarantee successful receipt of data. The experimental results show that this framework reduces data communication costs by nearly 37.15% and improves success rates by nearly 9.25%.

Exploring Scalable, Distributed Real-Time Anomaly Detection for Bridge Health Monitoring

Modern real-time Structural Health Monitoring systems can generate a considerable amount of information that must be processed and evaluated for detecting early anomalies and generating prompt warnings and alarms about the civil infrastructure conditions. The current cloud-based solutions cannot scale if the raw data has to be collected from thousands of buildings. This paper presents a full-stack deployment of an efficient and scalable anomaly detection pipeline for SHM systems which does not require sending raw data to the cloud but relies on edge computation. First, we benchmark three algorithmic approaches of anomaly detection, i.e., Principal Component Analysis (PCA), Fully-Connected AutoEncoder (FC-AE), and Convolutional AutoEncoder (C-AE). Then, we deploy them on an edge-sensor, the STM32L4, with limited computing capabilities. Our approach decreases network traffic by $\approx8\cdot10^5\times$ , from 780KB/hour to less than 10 Bytes/hour for a single installation and minimize network and cloud resource utilization, enabling the scaling of the monitoring infrastructure. A real-life case study, a highway bridge in Italy, demonstrates that combining near-sensor computation of anomaly detection algorithms, smart pre-processing, and low-power wide-area network protocols (LPWAN) we can greatly reduce data communication and cloud computing costs, while anomaly detection accuracy is not adversely affected.

An Efficient Distributed SPARQL Query Processing Scheme Considering Communication Costs in Spark Environments

Various distributed processing schemes were studied to efficiently utilize a large scale of RDF graph in semantic web services. This paper proposes a new distributed SPARQL query processing scheme considering communication costs in Spark environments to reduce I/O costs during SPARQL query processing. We divide a SPARQL query into several subqueries using a WHERE clause to process a query of an RDF graph stored in a distributed environment. The proposed scheme reduces data communication costs by grouping the divided subqueries in related nodes through the index and processing them, and the grouped subqueries calculate the cost of all possible query execution paths to select an efficient query execution path. The efficient query execution path is selected through the algorithm considering the data parsing cost of all possible query execution paths, amount of data communication, and queue time per node. It is shown through various performance evaluations that the proposed scheme outperforms the existing schemes.

On the Test Particle Monte-Carlo method to solve the steady state Boltzmann equation, the congruity of its results with experiments and its potential for shared memory parallelism

The Test Particle Monte Carlo is a known method to solve the steady state Boltzmann particle transport equation in rarefied gas systems. A description of the Test Particle Monte-Carlo procedure is outlined and the accuracy of this method is investigated by analyzing its consistency to experimental data of steady state effusions in isothermal systems. Computational results present deviations from the experiments that are at most 6.7%. After which, this paper investigates the potential of this method when it is parallelized using multi-core CPUs with shared memory for large Knudsen numbers. Scalability is expected since the method relies on a mean particle field that renders particle trajectories nearly independent from one another, therefore reducing data communication between processing cores. The mean particle distribution field helps linearize the collision term of the Boltzmann equation. Shared Memory Parallelism is an interesting feature when combined with distributed memory parallelism for design optimization of vacuum systems.

Minimizing Training Time of Distributed Machine Learning by Reducing Data Communication

Due to the additive property of most machine learning objective functions, the training can be distributed to multiple machines. Distributed machine learning is an efficient way to deal with the rapid growth of data volume at the cost of extra inter-machine communication. One common implementation is the parameter server system which contains two types of nodes: worker nodes, which are used for calculating updates, and server nodes, which are used for maintaining parameters. We observe that inefficient communication between workers and servers may slow down the system. Therefore, we propose a graph partition problem to partition data among workers and parameters among servers such that the total training time is minimized. Our problem is NP-Complete. We investigate a two-step heuristic approach that first partitions data, and then partitions parameters. We consider the trade-off between partition time and the saving in training time. Besides, we adopt a multilevel graph partition approach to fit the bipartite graph partitioning. We implement both approaches based on an open-source parameter server platform-PS-lite. Experiment results on synthetic and real-world datasets show that both approaches could significantly improve the communication efficiency up to 14 times compared with the random partition.

A less computationally complex clustering algorithm based on dynamic K-means for increasing lifetime of wireless sensor networks

Clustering in wireless sensor networks is a critical issue based on network lifetime, energy efficiency, connectivity and scalability. Sensor nodes are capable to collect data from any geographical region using routing protocol. This research endeavours to design a less computationally time complex clustering algorithm for hierarchical homogeneous wireless sensor network to extend network lifetime. It forms optimal number of clusters and reduces data communication span of sensor nodes using dynamic K-means algorithm. Selection of suitable cluster head is based on ratio of remaining energy of sensor node to its distance from centre of cluster. The simulation results prove that algorithm that has been presented achieves better energy efficiency when compared to other hierarchical homogeneous cluster based algorithms. It increases network lifetime, number of alive nodes per round, data delivered to base station, time of first node, middle node and last node to die for scalable situations in terms of node density and size of sensing region.

A Wearable Bio-signal Processing System with Ultra-low-power SoC and Collaborative Neural Network Classifier for Low Dimensional Data Communication.

In this paper, a real time physiological signal classification system with an integrated ultra-low power collaborative neural network classifier is presented. The developed system includes a specially designed system-on-chip (SoC) and a wireless communication module that transmits classification results to a smartphone app as a convenient user interface in real-time training. The customized SoC provides ultra-low-power and low-latency sensing and classification on physiological signals, e.g. EMG and ECG. A special collaborative neural network classifier was implemented to allow multiple chips to collaborate on classification. As a result, only low dimensional data is being transmitted over the network, significantly reducing data communication across multiple modules. A demonstration of EMG based gesture classification shows 1100X less power consumption from the developed SoC compared with conventional embedded solutions. The transmission of only low dimensional data from the collaborative neural network classifier leads to a 50X reduction of data communication and associated energy for multiple sensing cites.

Green content communications in 6LoWPAN

As a low-rate and low-power network, IPv6 over low-power wireless personal area network (6LoWPAN) wireless sensor network (WSN) suffers from considerable data communication latency and rapid energy depletion which data transmission mainly accounts for. Therefore, it is significant to reduce data communication latency and balance the energy consumption among sensor nodes to avoid the situation that a sensor node transmits a large content. To achieve these two objectives, the authors are motivated to extend the k -anycast communication model to 6LoWPAN WSN so that the different parts of a large content can be provided by multiple optimal k -anycast members in parallel. In this way, content communication latency can be reduced and energy consumption caused by content transmission can be shared by these members. This proposal is evaluated and the data results show that this proposal effectively reduces the content communication latency and achieves the energy consumption balance.

An improved algorithm for generalized least squares estimation

<p style='text-indent:20px;'>The textbook direct method for generalized least squares estimation was developed by Christopher C. Paige about 40 years ago. He proposed two algorithms. Suppose that the noise covariance matrix, rather than its factor, is available. Both of the Paige's algorithms involve three matrix factorizations. The first does not exploit the matrix structure of the problem, but it can be implemented by blocking techniques to reduce data communication time on modern computer processors. The second takes advantage of the matrix structure, but its main part cannot be implemented by blocking techniques. In this paper, we propose an improved algorithm. The new algorithm involves only two matrix factorizations, instead of three, and can be implemented by blocking techniques. We show that, in terms of flop counts, the improved algorithm costs less than Paige's first algorithm in any case and less than his second algorithm in some cases. Numerical tests show that in terms of CPU running time, our improved algorithm is faster than both of the existing algorithms when blocking techniques are used.

A new compression technique in MANET: compressed-LZW algorithm

&lt;p&gt;Now a days mobile ad hoc networks MANETs play a vital role in tremendous communication applications. This networks require infrastructure less module in non- centralized application’s area. Low battery power is of great concern to sustain for a long time, thus researchers try to assist in this critical maintenance point. This work proposed an enhanced version of well-known compression technique LZW to reduce data communication and maintain battery life. Compressed-LZW is a technique to code data stream into further compact form by merging successive redundant compressed code, and achieved by padding the code stream with the displacement of the next similar code. Compressed-LZW outperform its successor in saving consumed energy as expressed in the implementation and results by generating less output codes.&lt;/p&gt;

Energy- and locality-efficient multi-job scheduling based on MapReduce for heterogeneous datacenter

Job scheduling of MapReduce is a research hot spot, especially on the heterogeneous datacenter. Huge energy consumption and operating costs are key challenges. Most of the previous work only considers the scheduling optimization of a single job. In this paper, we take multiple jobs of MapReduce as research objects and focus on the goal of “jointly optimizing the scheduling time, job costs and energy consumption.” For that, an energy- and locality-efficient MapReduce multi-job scheduling algorithm is developed for the heterogeneous datacenter. Firstly, we use rack as the basic unit of resource in job scheduling to reduce data communication between jobs and to facilitate energy savings. Secondly, according to the capacity of heterogeneous rack, we design a multi-job pre-mapping method to optimize the execution order of jobs and jointly optimize the scheduling time, job costs and energy consumption. Based this pre-mapping method, we can assign one job to the virtual machine on the same rack, so as to minimize the amount of online rack. This centralized mapping strategy is very helpful to save energy and reduce data transmission of jobs. Thirdly, the map and reduce tasks of a job will be divided into multiple task groups for parallel execution, thereby further reducing data communication and energy consumption. Finally, a lot of experimental results prove the advantages of our algorithm.

Cloud-Edge Network Data Processing based on User Requirements using Modify MapReduce Algorithm and Machine Learning Techniques

Edge computing extends cloud computing to enhancing network performance in terms of latency and network traffic of many applications such as: The Internet of Things (IoT), Cyber-Physical Systems (CPS), Machine to Machine (M2M) technologies, Industrial Internet, and Smart Cities. This extension aims at reducing data communication and transmission through the network. However, data processing is the main challenge facing edge computing. In this paper, we proposed a data processing framework based on both edge computing and cloud computing, that is performed by partitioning (classification and restructuring) of data schema on the edge computing level based on feature selection. These features are detected using MapReduce algorithm and a proposed machine learning subsystem built on user requirements. Our approach mainly relies on the assumption that the data sent by edge devices can be used in two forms, as control data (i.e. real-time analytics) and as knowledge extraction data (i.e. historical analytics).We evaluated the proposed framework based on the amount of transmitted, stored data and data retrieval time, the results show that both the amount of sending data was optimized and data retrieval time was highly decreased. Our evaluation was applied experimentally and theoretically on a hypothetical system in a kidney disease center.