Local Computation Research Articles

Desktop as a Service using Local Computers

Abstract: Desktop-as-a-Service (DaaS) has gained prominence as a cloud-based solution for delivering desktop environments to users. However, traditional DaaS models often rely on remote data centers, posing concerns regarding data security, latency, and dependence on external infrastructure. In this paper, we propose a paradigm shift by introducing a novel approach to DaaS that leverages the untapped potential of local computing resources.Our system harnesses the idle computing power available within the user's local environment, including desktops, laptops, and servers, to create a distributed DaaS platform. By integrating virtualization technologies and dynamic resource allocation strategies, we enable the provisioning of desktop environments ondemand, effectively reducing the reliance on centralized data centers. Key components of our system include a centralized management platform for orchestrating resource allocation and workload distribution, virtualization techniques for isolating and managing desktop instances, and robust security measures to safeguard sensitive data. Through experimental validation, we demonstrate the feasibility and performance of our approach, showcasing improvements in data privacy, latency reduction, and resource utilization compared to traditional DaaS models. Our research presents a significant step towards decentralized and resilient computing ecosystems, offering users greater control over their desktop environments while maintaining the scalability and accessibility inherent to DaaS paradigms

An ameliorated localization algorithm for compensating stratification effect based on improved underwater salp swarm optimization technique

SummaryThe underwater acoustic sensor network is a fundamental source for ocean exploration. The potential applications of underwater acoustic sensor network (UASN) include seismic imaging, disaster prevention, mine reconnaissance, pollution monitoring, exploration of natural resources, and military surveillance. To acquire accurate results, implementing all applications of underwater sensor networks requires an adequate network connection and communication technology. The precise placement of underwater sensor nodes needs to be identified in order to achieve effective communication. To accomplish the requirement, the paper presents an efficient localization algorithm to compensate the stratification effect based on an improved underwater salp swarm optimization technique (LAS‐IUSSOT). To compute the location of sensor nodes with high accuracy, the nodes are initially randomly deployed in three‐dimensional underwater acoustic sensor network (3D‐UASN). After that, the hybridization of centroid‐based localization and the ray theory technique is used, and then, the degree of coplanarity is analyzed among the underwater sensor nodes. Then, the computation of the location of unknown nodes is performed using improved underwater salp swarm optimization technique (IUSSOT) to obtain the optimized location and compensate the impact of the stratification. The comparison of the simulation results of the existing algorithm and the proposed algorithm is performed. The LAS‐IUSSOT achieves 40.46% and 28.00% accuracy in terms of localization of underwater sensor nodes for both the sparse and dense regions in 3D‐UASN. The LAS‐IUSSOT achieves 49.39% and 62.57% accuracy in terms of ranging of underwater sensor nodes for both the sparse and dense regions in 3D‐UASN. Simulation results illustrate that the proposed algorithm outperforms the existing algorithm in terms of localization and ranging accuracy in both sparse and dense regions in 3D‐UASN, root mean square error (RMSE), normalized localization error (NLE), computation time, and convergence rate.

Primary intraorbital inflammatory lumpy lesion: A rare case report.

Eosinophilic angiocentric fibrosis (EAF) is considered to be a kind of benign IgG4-related disease, and it is more often found in the nasal cavity. We present a pretty rare case of orbital EAF that is unlike any other reported case for this case is an IgG4 negative orbital EAF and successfully treated by the fronto orbitozygomatic approach surgery. This is a 68-year-old man from a rural area of Inner Mongolia Autonomous Region, went to our hospital for a 2-month history of vision loss with a local hospital orbital computer tomography which showed that there was a lesion in his left orbit. The inspection of the patient revealed that the patient left eye was protruding outward and the left eyelid unable to complete open or close. And his left eyeball movement had difficulty in all directions. Postoperative pathology diagnosed that this was a case of IgG4-negative EAF case. Orbital EAF. Surgical radical resection and postoperative glucocorticoid therapy. After surgery, the left eye vision of this patient increased to 0.6 tested in the standard logarithmic visual acuity chart. And his left eyeball movement dysfunction and eyeball outward protruding get a partially relief. EAF occurring in the orbit is a very rare disease and immunohistochemical results of EAF can be IgG4 negative.

VR digital twin of office space with computer vision-based estimation of room occupancy and power consumption

AbstractIn the past years, energy consumption has increased rapidly due to many factors, including the rise in technology adoption. This has many downfalls, from higher costs to CO$$_2$$ 2 emissions. Human activities in offices and houses represent a considerable amount of energy usage. A digital twin (DT) of an open-plan common space is created, serving the purpose of remote room occupancy monitoring and automatic detection of energy consumption. A virtual reality (VR) model is developed and integrated to temperature, humidity and imaging sensors. For maintaining privacy, images are processed in local computers to measure occupancy levels and live video feed were never transmitted. The same set of imaging sensors were also used in a bespoke computer vision module for energy consumption estimation. The human avatars were mapped with high correlation (R$$^{2}$$ 2 $$= 0.85$$ = 0.85 ) with actual positions on floor. Our energy consumption algorithm accuracy obtained true positive rate of $$91.58\%$$ 91.58 % and F1 score of $$81.96\%$$ 81.96 % . Finally, all this information is transmitted and visualized to the 3D digital twin for remote monitoring and simulation.

Robust Offloading for Edge Computing-Assisted Sensing and Communication Systems: A Deep Reinforcement Learning Approach.

In this paper, we consider an integrated sensing, communication, and computation (ISCC) system to alleviate the spectrum congestion and computation burden problem. Specifically, while serving communication users, a base station (BS) actively engages in sensing targets and collaborates seamlessly with the edge server to concurrently process the acquired sensing data for efficient target recognition. A significant challenge in edge computing systems arises from the inherent uncertainty in computations, mainly stemming from the unpredictable complexity of tasks. With this consideration, we address the computation uncertainty by formulating a robust communication and computing resource allocation problem in ISCC systems. The primary goal of the system is to minimize total energy consumption while adhering to perception and delay constraints. This is achieved through the optimization of transmit beamforming, offloading ratio, and computing resource allocation, effectively managing the trade-offs between local execution and edge computing. To overcome this challenge, we employ a Markov decision process (MDP) in conjunction with the proximal policy optimization (PPO) algorithm, establishing an adaptive learning strategy. The proposed algorithm stands out for its rapid training speed, ensuring compliance with latency requirements for perception and computation in applications. Simulation results highlight its robustness and effectiveness within ISCC systems compared to baseline approaches.

After Mainframes: Computer Education and Microcomputers in Western Switzerland during the 1980s and 1990s

ABSTRACT This article analyses the developments in computer education in French-speaking Western Switzerland in the 1980s and 1990s. It investigates how computer education changed with the arrival of microcomputers, who the agents of these changes were, and with what sociopolitical and economic developments these pedagogical changes interacted. By analysing archival material from five cantonal departments of education and from Raymond Morel (at the time, Western Switzerland’s most influential educational IT expert), it is argued that computer education in Western Switzerland’s general education experienced three shifts. First, education with computers displaced education about computers in the first half of the 1980s. Second, from the mid-1980s onwards, office applications became central. Third, computer education morphed into telecommunications education around 1990. The paper contributes to the historiography of local school computers and computer education. It adds comment on the “glocal” histories of computers, which shed light on local computer cultures and economies across the globe.

EPIDL: Towards efficient and privacy‐preserving inference in deep learning

SummaryDeep learning has shown its great potential in real‐world applications. However, users(clients) who want to use deep learning applications need to send their data to the deep learning service provider (server), which can make the client's data leak to the server, resulting in serious privacy concerns. To address this issue, we propose a protocol named EPIDL to perform efficient and secure inference tasks on neural networks. This protocol enables the client and server to complete inference tasks by performing secure multi‐party computation (MPC) and the client's private data is kept secret from the server. The work in EPIDL can be summarized as follows: First, we optimized the convolution operation and matrix multiplication, such that the total communication can be reduced; Second, we proposed a new method for truncation following secure multiplication based on oblivious transfer and garbled circuits, which will not fail and can be executed together with the ReLU activation function; Finally, we replace complex activation function with MPC‐friendly approximation function. We implement our work in C++ and accelerate the local matrix computation with CUDA support. We evaluate the efficiency of EPIDL in privacy‐preserving deep learning inference tasks, such as the time to execute a secure inference on the MNIST dataset in the LeNet model is about 0.14 s. Compared with the state‐ofthe‐art work, our work is 1.8–98 faster over LAN and WAN, respectively. The experimental results show that our EPIDL is efficient and privacy‐preserving.

A radiometric and rotational invariant feature descriptor for multimodal remote sensing image matching using the monogenic signal

ABSTRACT Multimodal remote sensing image matching is still a challenging task, and its matching performance will highly influence the subsequent applications, such as image fusion, image mosaic and change detection. To tackle this problem, we propose a radiometric and rotational invariant descriptor (RRID) using the monogenic signal for multimodal image matching. First of all, a multiscale feature asymmetry measurement is proposed to obtain more reliable detected points, in which the multiscale feature asymmetry can be computed from the intermediate results from the monogenic signal; then a multiscale local phase orientation computation algorithm is designed for the main orientation assignment of each keypoint, in which the local phase orientation can also be provided by the monogenic signal construction; finally, a circular local window around each keypoint is extracted, a histogram of local phase orientations weighted by phase congruency is calculated for the log-polar feature description. RRID is compared with other state-of-the-art multimodal image matching methods on three publicly available datasets. Experimental results have demonstrated that our proposed method outperforms the compared methods in terms of precision, recall, and F1-measure criteria and also has achieved the lowest RMSEs.

Intelligent Reflecting Surface Assisted Secure Computation of Wireless Powered MEC System

The integration of mobile edge computing (MEC) and wireless power transfer (WPT) can effectively improve the computing ability and energy sustainability of energy-constrained wireless devices in the Internet of Things (IoT) networks. Intelligent reflecting surface (IRS) has recently emerged as an effective technique to improve the performance of wireless systems by intelligently reconfiguring wireless environments. This paper studies the exploitation of IRS to improve the secure computation performance of WPT-MEC systems with a passive eavesdropper. A wireless access point (AP) first charge multiple users with the emitted energy signals, and then the users perform local computing and partial offloading to complete their computation tasks with the harvested energy in the presence of an eavesdropper, where the local computing can be executed during the whole process of WPT and offloading. Meanwhile, deploying IRS can improve the energy capture and secure offloading performance of the users. We maximize the secure computation task bits of users by jointly optimizing the AP energy transmit beamforming, the IRS phase shifts, the transmit power, users' offloading time, and the local computation frequency of users, which are tangled with each other. An iterative optimal algorithm is developed to solve this non-convex problem by combining Taylor expansion method, semidefinite relaxation (SDR) algorithm, the Lagrange duality theory and Karush-Kuhn-Tucker (KKT) conditions. The numerical results show that the proposed scheme can effectively increase the secure computation task bits compared with other benchmark schemes, especially for the maximum transmit power of AP, the improvement is above 45 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\%$</tex-math></inline-formula> .

Embracing Distributed Systems for Efficient Cloud Resource Management: A Review of Techniques and Methodologies

The development of parallel computing, distributed computing, and grid computing has introduced a new computing model, combining elements of grid, public computing, and SaaS. Cloud computing, a key component of this model, assigns computing to distributed computers rather than local computers or remote servers. Research papers from 2017 to 2023 provide an overview of the advancements and challenges in cloud computing and distributed systems, focusing on resource management and the integration of advanced technologies like machine learning, AI-centric strategies, and fuzzy meta-heuristics. These studies aim to improve operational efficiency, scalability, and adaptability in cloud environments, focusing on energy efficiency and cost reduction. However, these advancements also present challenges, such as implementation complexity, adaptability in diverse environments, and the rapid pace of technological advancements. These issues necessitate practical, efficient, and forward-thinking solutions in real-world settings. The research conducted between 2017 and 2023 highlights the dynamic and rapidly evolving field of cloud computing and distributed systems, providing valuable guidance for ongoing and future research. This body of work serves as a crucial reference point for advancing the field and emphasizing the need for practical, efficient, and forward-thinking solutions in the ever-evolving landscape of cloud computing and distributed systems.

Energy Efficient and Differentially Private Federated Learning via a Piggyback Approach

This paper aims to develop a differential private federated learning (FL) scheme with the least artificial noises added while minimizing the energy consumption of participating mobile devices. By observing that some communication efficient FL approaches and even the nature of wireless communications contribute to the differential privacy (DP) preservation of training data on mobile devices, in this paper, we propose to jointly leverage gradient compression techniques (i.e., gradient quantization and sparsification) and additive white Gaussian noises (AWGN) in wireless channels to develop a piggyback DP approach for FL over mobile devices. Even with the piggyback DP approach, information distortion caused by gradient compression and noise perturbation may slow down FL convergence, which in turn consumes more energy of mobile devices for local computing and model update communications. Thus, we theoretically analyze FL convergence and formulate an energy efficient FL optimization under piggyback DP, transmission power, and FL convergence constraints. Furthermore, we propose an efficient iterative algorithm where closed-form solutions for artificial DP noise and power control are derived. Extensive simulation and experimental results demonstrate the effectiveness of the proposed scheme in terms of energy efficiency and privacy preservation.

Efficient Resource Management and Expansion Scheme for Collaborative Edge-Cloud Computing

Integrating the advantages of both the edge and the cloud, the edge-cloud computing system emerges to provide high-quality computing services for mobile users. To improve system efficiency, we investigate a hybrid mode of resource collaboration and expansion for the edge-cloud computing system, in which edge servers not only can collaborate with the cloud by purchasing high-priority computation resources temporarily but also can expand their local computation resources permanently. In such a way, the edge server can maximize its long-term profit by making a trade-off between the purchasing cost and the expanding cost. By formulating the resource management problem as a long-term profit maximization one, we first analyze the relationships among the expected minimal purchasing cost, the computation delay, and the available computation resources. Then, we design an efficient resource reserving and expanding scheme to determine the optimal expected amounts of reserving resources and expansion resources. Next, we propose an efficient real-time resource purchasing scheme to obtain the optimal amount of real-time purchasing resources dynamically. Finally, simulation results show that the proposed efficient resource collaboration and expanding scheme can maximize the long-term profit while guaranteeing the computation delay.

On the Role of Server Momentum in Federated Learning

Federated Averaging (FedAvg) is known to experience convergence issues when encountering significant clients system heterogeneity and data heterogeneity. Server momentum has been proposed as an effective mitigation. However, existing server momentum works are restrictive in the momentum formulation, do not properly schedule hyperparameters and focus only on system homogeneous settings, which leaves the role of server momentum still an under-explored problem. In this paper, we propose a general framework for server momentum, that (a) covers a large class of momentum schemes that are unexplored in federated learning (FL), (b) enables a popular stagewise hyperparameter scheduler, (c) allows heterogeneous and asynchronous local computing. We provide rigorous convergence analysis for the proposed framework. To our best knowledge, this is the first work that thoroughly analyzes the performances of server momentum with a hyperparameter scheduler and system heterogeneity. Extensive experiments validate the effectiveness of our proposed framework. Due to page limit, we leave all proofs to the full version https://arxiv.org/abs/2312.12670.

A locally calculable P3‐pressure in a decoupled method for incompressible Stokes equations

AbstractThis article will suggest a new finite element method to find a ‐velocity and a ‐pressure solving incompressible Stokes equations at low cost. The method solves first the decoupled equation for a ‐velocity. Then, using the calculated velocity, a locally calculable ‐pressure will be defined component‐wisely. The resulting ‐pressure is analyzed to have the optimal order of convergence. Since the pressure is calculated by local computation only, the chief time cost of the new method is on solving the decoupled equation for the ‐velocity. Besides, the method overcomes the problem of singular vertices or corners.

Hybrid learning of predictive mobile-edge computation offloading under differently-aged network states

By offloading computationally demanding applications to edge servers, mobile edge computing (MEC) can alleviate the stringent hardware requirements and save energy consumption of resource-restrained devices. Mobile edge computation offloading (MECO, i.e., optimizing computation offloading and resource allocation) is critical to the performance of MEC. However, the existing study typically assumed the availability of all the network states for the upcoming time slot, not applicable in practical MECO with differently-aged network states (e.g., fast-changing practical wireless channels and instantly observable local task arrivals). This paper proposes a novel hybrid learning approach that optimizes the instantaneous local processing and predictive computation offloading decisions. We establish a new hybrid learning approach for instantaneous local processing and predictive computation offloading decisions by integrating the learning techniques of stochastic gradient descent (SGD) and online convex optimization (OCO). By decomposing the primal problems, the proposed approach can be implemented in a decentralized manner. We prove that the optimality loss resulting from the differently-aged network states can diminish with the decreasing stepsizes of SGD and OCO. Simulation results validate the asymptotic optimality and superiority of the proposed hybrid learning approach, as compared to state-of-the-art (e.g., based on OCO techniques).

FedKNN: Secure Federated k-Nearest Neighbor Search

Nearest neighbor search is a fundamental task in various domains, such as federated learning, data mining, information retrieval, and biomedicine. With the increasing need to utilize data from different organizations while respecting privacy regulations, private data federation has emerged as a promising solution. However, it is costly to directly apply existing approaches to federated k-nearest neighbor (kNN) search with difficult-to-compute distance functions, like graph or sequence similarity. To address this challenge, we propose FedKNN, a system that supports secure federated kNN search queries with a wide range of similarity measurements. Our system is equipped with a new Distribution-Aware kNN (DANN) algorithm to minimize unnecessary local computations while protecting data privacy. We further develop DANN*, a secure version of DANN that satisfies differential obliviousness. Extensive evaluations show that FedKNN outperforms state-of-the-art solutions, achieving up to 4.8× improvement on federated graph kNN search and up to 2.7× improvement on federated sequence kNN search. Additionally, our approach offers a trade-off between privacy and efficiency, providing strong privacy guarantees with minimal overhead.

Machine learning for the prediction of local asteroid damages

Risk assessment studies of local asteroid hazards traditionally simulate the physics of meteors with engineering models tailored to analyze tens-of-millions of scenarios. However, these simplified approaches still need to solve time-dependent ODEs to model the entry process and the resulting ground damage. With a computational cost of O(0.01 CPU.s) per scenario, simulating these large numbers of potential entry conditions can take several days on local computers. To improve computational efficiency, we propose in this paper an orthogonal approach based on machine learning models to predict the size of damaged areas given a list of entry parameters. We train 5 machine learning methods and compare the predictions to the outputs of NASA’s state-of-the-art Probabilistic Asteroid Impact Risk (PAIR) model. PAIR’s remarkable capability to handle a broad spectrum of entry conditions and a high volume of scenarios establishes it as our reference model, with PAIR estimates assumed ground truth and used to train, test, and validate the machine learning models. The ML models, initially trained with PAIR’s primitive input variables, are then further enhanced with engineered features to improve the performance of interpretable models. We find that complex models like neural networks are well-suited to estimate blast hazards, while simpler, interpretable linear models can accurately assess thermal damage. For both types of hazards, the radii of damaged areas can be predicted with less than 10% average relative errors for radii greater than 100 km. The CPU time is decreased by a factor O(103) compared to the PAIR model, enabling the simulation of millions of scenarios within minutes on a local computer. Beyond serving as surrogate models, we incorporate the machine learning models into a comprehensive Shapley sensitivity analysis, yielding a ranking of entry parameters based on their contributions to ground damages. These results offer valuable insights for prioritizing observation missions.

Predicting protein retention in ion-exchange chromatography using an open source QSPR workflow.

Protein-based biopharmaceuticals require high purity before final formulation to ensure product safety, making process development time consuming. Implementation of computational approaches at the initial stages of process development offers a significant reduction in development efforts. By preselecting process conditions, experimental screening can be limited to only a subset. One such computational selection approach is the application of Quantitative Structure Property Relationship (QSPR) models that describe the properties exploited during purification. This work presents a novel open-source Python tool capable of extracting a range of features from protein 3D models on a local computer allowing total transparency of the calculations. As open-source tool, it also impacts initial investments in constructing a QSPR workflow for protein property prediction for third parties, making it widely applicable within the field of bioprocess development. The focus of current calculated molecular features is projection onto the protein surface by constructing surface grid representations. Linear regression models were trained with the calculated features to predict chromatographic retention times/volumes. Model validation shows a high accuracy for anion and cation exchange chromatography data (cross-validated R2 of 0.87 and 0.95). Hence, these models demonstrate the potential of the use of QSPR to accelerate process design.

Intelligent task processing using mobile edge computing: processing time optimization

&lt;p&gt;The fast-paced development of the internet of things led to the increase of computing resource services that could provide a fast response time, which is an unsatisfied feature when using cloud infrastructures due to network latency. Therefore, mobile edge computing became an emerging model by extending computation and storage resources to the network edge, to meet the demands of delaysensitive and heavy computing applications. Computation offloading is the main feature that makes Edge computing surpass the existing cloud-based technologies to break limitations such as computing capabilities, battery resources, and storage availability, it enhances the durability and performance of mobile devices by offloading local intensive computation tasks to edge servers. However, the optimal solution is not always guaranteed by offloading computation, therefore, the offloading decision is a crucial step depending on many parameters that should be taken in consideration. In this paper, we use a simulator to compare a two tier edge orchestrator architecture with the results obtained by implementing a system model that aims to minimize a task’s processing time constrained by time delay and the limited device’s computational resource and usage based on a modified version.&lt;/p&gt;

Stochastic Resource Optimization for Wireless Powered Hybrid Coded Edge Computing Networks

With the ubiquitous sensing enabled by the Internet-of-Things (IoT), massive amount of data is generated every second, transforming the way we interact with the world. To manage big data and enable analytics at the edge of the network, large amount of computation power is required to perform the computation intensive tasks. However, the energy-constrained IoT devices are not able to perform the computation tasks without compromising the quality-of-service of the applications. In this paper, we propose a hybrid network in which users can fully offload their computation tasks to edge servers through coded edge offloading or perform local computation with the wireless power transfer derived from coalitions of unmanned aerial vehicles (UAVs) serving as mobile charging stations. To minimize the cost of the network, we consider a two-level optimization approach. At the lower level, an optimal UAV coalition that minimizes the network cost is formed. At the upper level, the computation approach for each user is decided while taking into account the stochastic nature of wireless charging efficiency. Given the interrelation between the two levels, we adopt the backward induction optimization strategy to jointly derive the optimal coalition structure and computation approach for each user. In the performance evaluation, we provide extensive sensitivity analyses to study the performance of the cost minimization approaches amid varying network parameters. Moreover, we show that our approach outperforms deterministic optimization approaches in network cost minimization.