Distributed Computing Resources Research Articles

The Role of Fog Device Density in IoT-Fog-Cloud Systems

Fog computing has emerged as an essential technology for enabling real-time, low-latency responses in cloud-based applications within Internet of Things (IoT) systems. This study explored the impact of the number of fog devices (NFDs) on the performance of IoT-fog-cloud systems. Through comparative analysis, two scheduling algorithms—First Come First Serve (FCFS) and Priority—were evaluated across different scales of system deployment. The results indicated that the FCFS algorithm was optimal for systems with fewer NFDs, whereas the Priority algorithm proved advantageous in larger settings. These findings not only establish guidelines for selecting appropriate scheduling strategies based on the scale of fog device deployment but also provide strategic insights for businesses on the optimal distribution of computational resources. This research aids companies in deciding whether to centralize resources in fewer branches with more employees or to decentralize into more branches with fewer employees, thereby optimizing operational efficiency and responsiveness.

Energy-Efficient Federated Learning Over Cell-Free IoT Networks: Modeling and Optimization

To leverage massive distributed data and computation resources in the Internet-of-Things (IoT) networks, federated learning (FL) is considered to be a promising technique with benefits of improved data privacy and communication efficiency. Meanwhile, cell-free massive multiple-input multiple-output (cell-free massive MIMO) is a promising technology to enable the IoT networks to support FL. By deploying the access points (APs) closer to the IoT devices, path loss attenuation can be reduced. However, the performance of FL is still constrained by the limited power resources of IoT devices. To address this issue, we design an energy-efficient FL scheme over cell-free IoT networks by formulating an optimization problem to minimize the total energy consumption of the IoT devices participating in the FL process. To solve the intractable problem in hand, by exploiting its unique structure, we decompose it into three sub-problems that facilitate the development of the proposed scheme. First, we derive the optimal central processing unit (CPU) operating frequency for IoT devices. Then, we design an optimal power allocation scheme to mitigate the straggler effect. Next, nonlinear programming method is adopted to obtain a sub-optimal solution for the reformulated problem. Finally, a three-stage algorithm is proposed for energy consumption minimization by considering these sub-problems. Simulation results demonstrate the close-to-optimal performance of the proposed algorithm for energy savings compared with three baseline algorithms and the capability to support large numbers of IoT device access by mitigating the straggler effect.

Federated Learning Over Wireless IoT Networks With Optimized Communication and Resources

To leverage massive distributed data and computation resources, machine learning in the network edge is considered to be a promising technique, especially for large-scale model training. Federated learning (FL), as a paradigm of collaborative learning techniques, has obtained increasing research attention with the benefits of communication efficiency and improved data privacy. Due to the lossy communication channels and limited communication resources (e.g., bandwidth and power), it is of interest to investigate fast responding and accurate FL schemes over wireless systems. Hence, we investigate the problem of jointly optimized communication efficiency and resources for FL over wireless Internet of Things (IoT) networks. To reduce complexity, we divide the overall optimization problem into two subproblems, i.e., the client scheduling problem and the resource allocation problem. To reduce the communication costs for FL in wireless IoT networks, a new client scheduling policy is proposed by reusing stale local model parameters. To maximize successful information exchange over networks, a Lagrange multiplier method is first leveraged by decoupling variables, including power variables, bandwidth variables, and transmission indicators. Then, a linear-search-based power and bandwidth allocation method is developed. Given appropriate hyperparameters, we show that the proposed communication-efficient FL (CEFL) framework converges at a strong linear rate. Through extensive experiments, it is revealed that the proposed CEFL framework substantially boosts both the communication efficiency and learning performance of both training loss and test accuracy for FL over wireless IoT networks compared to a basic FL approach with uniform resource allocation.

Разработка программного обеспечения электропривода для группового управления в электромеханической системе

Background. To operate multi-coordinate electromechanical systems require control of electric drives coordinated in time and space along each of the axes, which is called coordinated control. Thus, various computing tools and mathematical algorithms are used. They establish the baseline of the software for microprocessor control systems. At the same time, the implementation of control algorithms is limited by the computing resources of the hardware parts of control systems, including the limitations of the tools of information transmission. To consider and eliminate these limitations and thus, to improve operation of the multi-coordinate electromechanical system, the aim of the article is to update control algorithms and balance distribution of computational resources of electric drives control systems. Materials and methods. Data are presented on the software development for coordinated control of several electric drives as a part of an electromechanical system that performs coordinated motion along its axes in time and space. To synchronize the control, the following technical solutions have been applied: description of the path of the high-level controller in the form of a sequence of reference points, buffering the coordinates when received by the electric drives and subsequent interpolation. Balanced distribution of resources between the high-level controller and local systems of electric drives is carried out considering the intensity of data exchange via the network interface. When calculating the reference points by the high-level controller, approximation of the paths by line segments at time intervals of variable duration is carried out to describe the generic path within the allowable mismatch. Results. The algorithm to synchronize the system time of the electric drives and the high-level controller has been implemented to synchronize the interpolation of paths. To implement this algorithm and transfer the coordinates of reference points to electric drives, special commands are used. They have different priority levels depending on the required speed of execution. In accordance with this algorithm, software has been developed as an element of the microprocessor control system of the electric drive. Conclusions. Application of various technical tools and computational algorithms that make it possible to allocate the computing resources of the control system and rationally use them contributes to efficiency improvement, in particular, the consistency and timeliness of electric drives control as a part of multi-axis electromechanical systems. The authors have proved the efficiency of the approach of coordinated control with the use of buffering coordinates of reference points followed by interpolation of the paths depending on the system time of each of the electric drives.

ЦЕНТРАЛЬНО-КОЛЬЦЕВОЙ ПОЛИНОМИАЛЬНЫЙ АЛГОРИТМ РАСПРЕДЕЛЕНИЯ ВЫЧИСЛИТЕЛЬНО-ВРЕМЕННЫХ РЕСУРСОВ В ГРИД-СИСТЕМАХ

The article is devoted to solving the problem of computational and time resources distributionin grid systems based on the adaptation of polynomial algorithms to quadratic types of user applications.The relevance of demand distribution validity problem for the distributed computing paradigmin the context of information redistribution and uncertainty. The article deals with the problems ofscheduling heterogeneous computing resources in solving complex professional and scientific problemsachieved at different points in time, based on identifying resources by significant manifestationsof commitment and probability. A comparative review of consumption has been carried out. Thestatement of the problem to be solved in the chosen research area is formulated. The problem ofscheduling a grid system with a centralized multiarchitecture, which uses the task solution of agroup-site, is substantiated. The use of this architecture requires the development of heuristic algorithmsfor the distribution of computing resources, taking into account the properties of applicationarrays and assessing the schedule compliance. Eliminating the occurrence of scheduling errors requiresthe development of a formal apparatus that will identify the prospects of the application, introducetheir typing and build heuristic algorithms with quality assessment, selected for certain types.The development of such a formal apparatus is an urgent task. An equally important task within theframework of this mechanism is the construction of resource parity models and interaction betweenusers and the computing system models. The authors proposed to solve the problem of schedulingcomputing resources based on the development and study of polynomial scheduling algorithms forarrays of hyperbolic applications. The main theoretical accuracy of this study is the creation of aformal scheduling apparatus, including the definition of resource sugar, as a model of user applications,based on the performance of an operation in the scheduling environment on a set of resourcemuscles. The scientific novelty of the research lies in the development of a central-ring polynomialalgorithm for the distribution of computational time resources in grid systems, which involves anautomatic scheduling algorithm for computing systems, adaptation to quadratic types of user applicationsand improves the efficiency of computational time resources distribution. To evaluate thedeveloped efficiency of the software application algorithm and the conducted computational experimentwith rapidly generated classes of computational resources. Obtained comparative results of theproposed algorithm practical efficiency experimental studies for the distribution of computationaland time resources. The described studies have a high level of theoretical and practical significanceand are directly related to the solution of artificial intelligence classical problems.

Optimal Cellular Microscopic Pattern Recognizer- (OCMPR-) Based Wireless Detection Network for Efficiently Leveraging the Parallel Distributed Processing Capabilities.

Recognizing patterns associated with particular events enables the detection of specific critical changes in the events. Due to the resource constraints inherent in WSNs, pattern recognition is highly dependent on the complexity of the computation, the number of iterations, and the requirements for node training. Iterative learning is frequently used in computer-based computer vision. As a result, these methods are in conflict with the perfectly alright architecture of the WSN. The proposed technique, Optimal Cellular Microscopic Pattern Recognizer (OCMPR), enables the detection of macroscale events in WSN. Using the distributed system computational resources of WSNs, the approach reduces calculations for conserving energy and improves recognition. The method generates promising results by combining a well-known optimization technique (the genetic algorithm) with CMPR. This approach addresses the resource-constrained WSN's real-time mission-critical application needs. Global and quick recognition is achieved by dispersing processing over a network's nodes, allowing for loosely connected communication. The results demonstrate the suggested scheme's versatility.

Adaptive Subcarrier, Parameter, and Power Allocation for Partitioned Edge Learning Over Broadband Channels

In this paper, we consider <i>partitioned edge learning</i> (PARTEL), which implements parameter-server training, a well known distributed learning method, in a wireless network. Thereby, PARTEL leverages distributed computation resources at edge devices to train a large-scale <i>artificial intelligence</i> (AI) model by dynamically partitioning the model into parametric blocks for separated updating at devices. Targeting broadband channels, we consider the joint control of parameter allocation, sub-channel allocation, and transmission power to improve the performance of PARTEL. Specifically, the policies for joint SUbcarrier, Parameter, and POweR allocaTion (SUPPORT) are optimized under the criterion of minimum learning latency. Two cases are considered. First, for the case of decomposable models (e.g., logistic regression), the latency-minimization problem is a mixed-integer program and non-convex. Due to its intractability, we develop a practical solution by integer relaxation and transforming it into an equivalent convex problem of model size maximization under a latency constraint. Thereby, a low-complexity algorithm is designed to compute the SUPPORT policy. Second, consider the case of <i>deep neural network</i> (DNN) models which can be trained using PARTEL by introducing some auxiliary variables. This, however, introduces constraints on model partitioning reducing the granularity of parameter allocation. The preceding policy is extended to DNN models by applying the proposed techniques of load rounding and proportional adjustment to rein in latency expansion caused by the load granularity constraints.

Full spectrum optical constant interface to the Materials Project

Optical constants characterize the interaction of materials with light and are important properties in material design. Here we present a Python-based Corvus workflow for simulations of full spectrum optical constants from the visible and ultraviolet to hard x-ray wavelengths based on the real-space Green’s function code FEFF10 and structural data from the Materials Project (MP). The Corvus workflow manager and its associated tools provide an interface to FEFF10 and the MP database. The workflow parallelizes the FEFF computations of optical constants over all absorption edges for each material in the MP database specified by a unique MP-ID. The workflow tools determine the distribution of computational resources needed for that case. Similarly, the optical constants for selected sets of materials can be computed in a single-shot. To illustrate the approach, we present results for several elemental solids in the periodic table, as well as a sample compound, and compare our predictions with experimental results. In addition, we provide a database of calculated results for all elements for which there is a stable elemental solid at standard conditions available in the Materials Project database. As in x-ray absorption spectra, these results are interpreted in terms of an atomic-like background and fine-structure contributions.

On the Design of Federated Learning in the Mobile Edge Computing Systems

The combination of artificial intelligence and mobile edge computing (MEC) is considered as a promising evolution path of the future wireless networks. As a model-level coordination learning paradigm, federated learning can make full use of the distributed computation resource in the MEC systems, which allows the users to keep their private data locally. However, due to the unreliable wireless transmission circumstances and resource constraints in the MEC systems, both the performance and training efficiency of federated learning cannot be guaranteed. To solve this problem, the optimization design of federated learning in the MEC systems is studied in this paper. First, an optimization problem is formulated to manage the tradeoff between model accuracy and training cost. Second, a joint optimization algorithm is designed to optimize the model compression, sample selection, and user selection strategies, which can approach a stationary optimal solution in a computationally efficient way. Finally, the performance of our proposed optimization scheme is evaluated by numerical simulation and experiment results, which show that both the accuracy loss and the cost of federated learning in the MEC systems can be reduced significantly by employing our proposed algorithm.

An Evolutionary Algorithm for Task scheduling Problem in the Cloud-Fog environment

The rapid and enormous growth of the Internet of Things, as well as its widespread adoption, has resulted in the production of massive quantities of data that must be processed and sent to the cloud, but the delay in processing the data and the time it takes to send it to the cloud has resulted in the emergence of fog, a new generation of cloud in which the fog serves as an extension of cloud services at the edge of the network, reducing latency and traffic. The distribution of computational resources to minimize makespan and running costs is one of the disadvantages of fog computing. This paper provides a new approach for improving the task scheduling problem in a Cloud-Fog environment in terms of execution time(makespan) and operating costs for Bag-of-Tasks applications. A task scheduling evolutionary algorithm has been proposed. A single custom representation of the problem and a uniform intersection are built for the proposed algorithm. Furthermore, the individual initialization and perturbation operators (crossover and mutation) were created to resolve the inapplicability of any solution found or reached by the proposed evolutionary algorithm. The proposed ETS (Evolutionary Task Scheduling algorithm) algorithm was evaluated on 11 datasets of varying size in a number of tasks. The ETS outperformed the Bee Life (BLA), Modified Particle Swarm (MPSO), and RR algorithms in terms of Makespan and operating costs, according to the results of the experiments.

Multi-analysis surveillance and dynamic distribution of computational resources: Towards extensible, robust, and efficient monitoring of environments

Intelligent surveillance has been a hot area of research for the past two decades. However, the integral security of environments remains a challenge due to the presence of new types of threats and the very dynamism and complexity of these environments. To make further progress, new proposals are needed to facilitate the design and deployment of multi-analysis surveillance systems where different types of analysis are simultaneously performed and the available computation resources are limited. Such systems need to provide three main characteristics: extensibility, robustness and efficiency. Extensibility to add new analysis components when new events of interest must be monitored. Robustness to avoid that failures in one analysis component affect the rest. Efficiency to analyze environments in real-time and to support decision-making processes that address the detected anomalies.This paper proposes a formal model for the multi-analysis surveillance of environments by means of the named components of normality, designed to deploy surveillance systems that satisfies the three previously mentioned characteristics. Extensibility thanks to the activation and deactivation of components of normality based on the monitoring needs of the analyzed environment. Robustness as a result of the isolation of these components such that a failure in one of them does not spread to the rest. Finally, efficiency due to the dynamic allocation of resources that benefits the components that detect anomalies at a given time. The experimental results prove that such dynamic allocation improves execution times and reduces waiting times.

Resource Allocation for Edge Computing-Based Vehicle Platoon on Freeway: A Contract-Optimization Approach

Vehicular edge computing is viewed as a promising technique for relieving the overload of base station (BS) on roadside, via leveraging the computation resources of vehicles on road. However, the computing capacity of a single vehicle is limited. For the fast velocity and distribution characteristics of vehicles, when a vehicle performs task offloading to a set of neighbor vehicles, longer sensing/accessing time durations are needed. Vehicle platoon driving as an emerging technique is proposed to strengthen vehicle fuel efficiency and the freeway safety and capacity. The distributed computation resources can be centrally controlled in the vehicle platoon. In this paper, we investigate the resource allocation scheme for edge computing-based vehicle platoon on freeway to maximize the utilities of vehicles surrounding the platoon via a contract-optimization approach. Combining with BS, the vehicle platoon performs as an edge node to provide the delay-sensitive computation service for the vehicles on road. First, we design a contract-based incentive mechanism to motivate vehicles to form a vehicle platoon and share their computation resources. Then, based on the proposed vehicle platoon, a joint optimization problem is formulated to maximize the utilities of vehicles, the task offloading decision, communication and computation resource allocation schemes in both the platoon and BS are jointly optimized. Numerical results reveal that the designed contracts can incentivize vehicle's participation, and the proposed resource allocation schemes are effective.

Data Processing Model to Perform Big Data Analytics in Hybrid Infrastructures

Big Data applications are present in many areas such as financial markets, search engines, stream services, health care, social networks, and so on. Data analysis provides value to information for organizations. Classical Cloud Computing represents a robust architecture to perform complex and large-scale computing for these areas. The main challenges are the user's unknowledge about Cloud infrastructure, the requirement needed for improving performance, and the resource management to maintain stable processing. In these difficulties, an inadequate solution can lead to users overestimate or underestimate the number of computational resources, which drives to the budget increases. One way to work around this problem is to make use of Volunteer Computing since it provides distributed computational resources at free monetary cost. However, a volatile machine behavior is a problem to address in Big Data data distributions. Thus, this work proposes a data distribution model composed of Cloud Computing and Volunteer Computing environments in a hybrid fashion for Big Data analytics. The contributions of this work are: i) the required evaluation to enable efficient deployment of Big Data in hybrid infrastructures; ii) the development of an HR_Alloc Algorithm for establishing the data placement to Big Data applications; iii) a model to resource allocation in hybrid infrastructures. The obtained results indicate the feasibility of using a hybrid infrastructure with up to 35% of unstable machines in the worst-case scenario, without losing performance and a monetary cost lower than 20% in comparison to Classical Cloud Computing. Also, communication costs decrease up to 57.14% in the best-case scenario due to load balancing.

Cooperative Mining in Blockchain Networks With Zero-Determinant Strategies.

In proof-of-work (PoW)-based blockchain networks, the miners contribute their distributed computation in solving a crypto-puzzle competition to win the reward. To secure stable profits, some miners organize mining pools and share the rewards from the pool in proportion to each miner's contribution. However, some miners may exhibit malicious behaviors which cause a waste of distributed computation resource, even posing a threat on the efficiency of blockchain networks. In this paper, we propose a new game-theoretic framework to incentivize miners mining honestly and help to bring about a higher total welfare of blockchain networks. We first formulate the mining process as a noncooperative iterated game. We then propose a mechanism in terms of zero-determinant strategies (ZD strategies) to encourage the cooperative mining and improve the efficiency of mining in PoW-based blockchain networks. In addition, we theoretically analyze the maximum system welfare of the target pool through the method of optimization. Numerical illustrations are also presented to support our theoretical results.

Incentivizing Honest Mining in Blockchain Networks: A Reputation Approach

The core security of proof-of-work (PoW)-based blockchain networks, relies on PoW consensus algorithm and requires miners solving a crypto-puzzles of hash computation. However, the mining process in the blockchain is resource-intensive where only the first miner who completes full PoW will be rewarded. Therefore, miners exhibit malicious behaviors which cause a waste of distributed computation resource, even posing a threat on the efficiency of blockchain networks. In this brief, we propose a new reputation-based mechanism for the PoW computation in the blockchain, in which miners are incentivized to conduct honest mining. Based on the game theory, we design a reputation-based algorithm to encourage honest mining of miners, and thereby increase the overall revenue of the pool. In addition, numerical illustrations are also presented to support the performance of our proposed mechanism.

Efficient brownfield optimization of a reservoir in west Siberia

In this work we present a methodology for optimal management of brownfields that is illustrated on a real field. The approach does not depend on the particular reservoir flow simulator used, although streamline-derived information is leveraged to accelerate the optimization. The method allows one to include (non-linear) constraints (e.g. a recovery factor larger than a given baseline value), which are very often challenging to address with optimization tools. We rely on derivative-free optimization coupled with the filter method for non-linear constraints, although the methodology can also be combined with approaches that utilize exact/approximate gradients. Performance in terms of wall-clock time can be improved further if distributed-computing resources are available (the method is amenable to parallel implementation). The methodology is showcased using a real field in west Siberia where net present value (NPV) is maximized subject to a constraint for the recovery factor. The optimization variables represent a discrete time series for well bottom-hole pressure over a fraction of the production time frame. An increase in NPV of 7.9% is obtained with respect to an existing baseline. The optimization methods studied include local optimization algorithms (e.g. generalized pattern search) and global search procedures (e.g. particle swarm optimization). The controls for one injection well in the real field were actually modified according to the solution determined in this work. The results obtained suggest improvement for most economic scenarios.

Firework: Data Processing and Sharing for Hybrid Cloud-Edge Analytics

Now we are entering the era of the Internet of Everything (IoE) and billions of sensors and actuators are connected to the network. As one of the most sophisticated IoE applications, real-time video analytics is promising to significantly improve public safety, business intelligence, and healthcare & life science, among others. However, cloud-centric video analytics requires that all video data must be preloaded to a centralized cluster or the cloud, which suffers from high response latency and high cost of data transmission, given the scale of zettabytes of video data generated by IoE devices. Moreover, video data is rarely shared among multiple stakeholders due to various concerns, which restricts the practical deployment of video analytics that takes advantages of many data sources to make smart decisions. Furthermore, there is no efficient programming interface for developers and users to easily program and deploy IoE applications across geographically distributed computation resources. In this paper, we present a new computing framework, <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Firework</i> , which facilitates distributed data processing and sharing for IoE applications via a virtual shared data view and service composition. We designed an easy-to-use programming interface for <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Firework</i> to allow developers to program on <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Firework</i> . This paper describes the system design, implementation, and programming interface of Firework. The experimental results of a video analytics application demonstrate that <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Firework</i> reduces up to 19.52 percent of response latency and at least 72.77 percent of network bandwidth cost, compared to a cloud-centric solution.

A Holistic Approach for Collaborative Workload Execution in Volunteer Clouds

The demand for provisioning, using, and maintaining distributed computational resources is growing hand in hand with the quest for ubiquitous services. Centralized infrastructures such as cloud computing systems provide suitable solutions for many applications, but their scalability could be limited in some scenarios, such as in the case of latency-dependent applications. The volunteer cloud paradigm aims at overcoming this limitation by encouraging clients to offer their own spare, perhaps unused, computational resources. Volunteer clouds are thus complex, large-scale, dynamic systems that demand for self-adaptive capabilities to offer effective services, as well as modeling and analysis techniques to predict their behavior. In this article, we propose a novel holistic approach for volunteer clouds supporting collaborative task execution services able to improve the quality of service of compute-intensive workloads. We instantiate our approach by extending a recently proposed ant colony optimization algorithm for distributed task execution with a workload-based partitioning of the overlay network of the volunteer cloud. Finally, we evaluate our approach using simulation-based statistical analysis techniques on a workload benchmark provided by Google. Our results show that the proposed approach outperforms some traditional distributed task scheduling algorithms in the presence of compute-intensive workloads.

HIGH TECHNOLOGY SOFTWARE WEB-TOOLS TO SOLVE ENVIRONMENTAL PROBLEMS OF COAL REGION

The paper is dedicated to the pilot system of the computer information portal, that is being designed in Kemerovo State University in order to enable engineers, students, postgraduate students and other users to get an expanded access to solutions of environmental applied problems in Kuzbass. The following elements are considered to be pilot system elements: solution of the grit motion in a klooded shaft problem, virtual laboratory of parallel programming, distributed computer resources access system. The present research is based on the task № 2014/64 of state research “Scientific research organization”

Data-aware optimization of bioinformatics workflows in hybrid clouds

Life Sciences have been established and widely accepted as a foremost Big Data discipline; as such they are a constant source of the most computationally challenging problems. In order to provide efficient solutions, the community is turning towards scalable approaches such as the utilization of cloud resources in addition to any existing local computational infrastructures. Although bioinformatics workflows are generally amenable to parallelization, the challenges involved are however not only computationally, but also data intensive. In this paper we propose a data management methodology for achieving parallelism in bioinformatics workflows, while simultaneously minimizing data-interdependent file transfers. We combine our methodology with a novel two-stage scheduling approach capable of performing load estimation and balancing across and within heterogeneous distributed computational resources. Beyond an exhaustive experimentation regime to validate the scalability and speed-up of our approach, we compare it against a state-of-the-art high performance computing framework and showcase its time and cost advantages.