Memory Usage Research Articles

ANALISIS PENGGUNAAN WSL, VMWARE, DAN VIRTUAL BOX DI ATAS SISTEM OPERASI WINDOWS

This research focuses on the analysis of usage between Windows subsystem for Linux (WSL), VMware, and VirtualBox when run on top of the Windows operasi system. The main purpose of this study is to determine which platform offers the best performance in various usage scenarios. In this study, various factors such as performance, compatibility, memory usage, and device compatibility are examined. The results of this study indicate that performance between these three platforms varies depending on usage scenarios and system configurations. This research not only provides practical insights for end users looking for the most suitable virtualization platform for their needs, but also contributes to research literature in this field. The research findings can be a valuable reference for other researchers interested in further pursuing research on the usage between WSL, VMware and VirtualBox. Thus, this research is expected to provide valuable insights for both users and developers who are considering using WSL, VMware or VirtualBox for their needs and is expected to enrich our understanding of effective use of virtualization technologies. By considering various aspects, this study attempts to provide a clearer and more comprehensive picture of the performance and efficiency of these three platforms between WSL, VMware and VirtualBox in various usage scenarios. In addition, this research can also be a valuable reference for other researchers interested in continuing research in the field of usage between WSL, VMware and VirtualBox.

DDoS detection in electric vehicle charging stations: A deep learning perspective via CICEV2023 dataset

Distributed Denial of Service (DDoS) attacks have always been an important research topic in the field of information security. Regarding specialized infrastructures such as electric vehicle charging stations, detecting and preventing such attacks becomes even more critical. In the existing literature, most studies on DDoS attack detection focus on traditional methods that analyze network metrics such as network traffic, packet rates, and number of connections. These approaches attempt to detect attacks by identifying anomalies and irregularities in the network, but can have high error rates and fail to identify advanced attacks. Conversely though, detection methods based on system metrics use deeper and more insightful parameters such as processor utilization, memory usage, disk I/O operations, and system behavior. Such metrics provide a more detailed perspective than network-based approaches, allowing for more accurate detection of attacks. However, work in this area is not yet widespread enough further research and improvement are needed. The adoption of advanced system metrics-based methods can significantly improve the effectiveness of DDoS defense strategies, especially in next-generation and specialized infrastructures. This paper evaluates the applicability and effectiveness of Long Short-Term Memory (LSTM) and Feed-Forward Network (FFN) in detecting DDoS attacks against electric vehicle charging stations through system metrics using CICEV2023 dataset. Experimental results show that the LSTM based model offers advantages in terms of speed and processing capacity, while the FFN is superior in terms of the accuracy.

A Novel Blockchain-Based Privacy-Preserved Data Sharing Using Deep Maxout Network in Federated Learning

The quick emergence in the quantity of data produced through the linked devices of Internet of Things (IoT) models opened the novel potential to improve service qualities for budding tools considering data sharing. However, privacy problems are main issues of data providers for sharing data. The outflow of confidential data causes severe problems beyond the loss in finance of providers. A blockchain-based secured data-sharing model is devised for dealing with various kinds of parties. Thus, data-sharing issue is modeled as a machine learning issue by adapting federated learning (FL). Here, data privacy is controlled by sharing data in spite of exposing genuine data. At last, the FL is combined in consensus task of permissioned blockchain for accomplishing federated training. Here, the data model learning is executed using a deep maxout network (DMN), which is trained using jellyfish search African vultures optimization (JSAVO). Moreover, the data-sharing records are generated to share data amid data providers and requestors. The proposed JSAVO-based DMN outperformed with better accuracy of 93.3%, FPR of 0.054, loss function of 0.067, mean square error (MSE) of 0.346, mean average precision of 94.6, RMSE of 0.589, computational time of 17.47[Formula: see text]s, and memory usage of 48.62[Formula: see text]MB.

Real-Time Dense Visual SLAM with Neural Factor Representation

Developing a high-quality, real-time, dense visual SLAM system poses a significant challenge in the field of computer vision. NeRF introduces neural implicit representation, marking a notable advancement in visual SLAM research. However, existing neural implicit SLAM methods suffer from long runtimes and face challenges when modeling complex structures in scenes. In this paper, we propose a neural implicit dense visual SLAM method that enables high-quality real-time reconstruction even on a desktop PC. Firstly, we propose a novel neural scene representation, encoding the geometry and appearance information of the scene as a combination of the basis and coefficient factors. This representation allows for efficient memory usage and the accurate modeling of high-frequency detail regions. Secondly, we introduce feature integration rendering to significantly improve rendering speed while maintaining the quality of color rendering. Extensive experiments on synthetic and real-world datasets demonstrate that our method achieves an average improvement of more than 60% for Depth L1 and ATE RMSE compared to existing state-of-the-art methods when running at 9.8 Hz on a desktop PC with a 3.20 GHz Intel Core i9-12900K CPU and a single NVIDIA RTX 3090 GPU. This remarkable advancement highlights the crucial importance of our approach in the field of dense visual SLAM.

Memory-efficient DRASiW Models

Weightless Neural Networks (WNN) are ideal for Federated Learning due to their robustness and computational efficiency. These scenarios require models with a small memory footprint and the ability to aggregate knowledge from multiple models. In this work, we demonstrate the effectiveness of using Bloom filter variations to implement DRASiW models—an adaptation of WNN that records both the presence and frequency of patterns—with minimized memory usage. Across various datasets, DRASiW models show competitive performance compared to models like Random Forest, k-Nearest Neighbors, Multi-layer Perceptron, and Support Vector Machines, with an acceptable space trade-off. Furthermore, our findings indicate that Bloom filter variations, such as Count Min Sketch, can reduce the memory footprint of DRASiW models by up to 27% while maintaining performance and enabling distributed and federated learning strategies.

Learning the feature distribution similarities for online time series anomaly detection

Identifying anomalies in multi-dimensional sequential data is crucial for ensuring optimal performance across various domains and in large-scale systems. Traditional contrastive methods utilize feature similarity between different features extracted from multidimensional raw inputs as an indicator of anomaly severity. However, the complex objective functions and meticulously designed modules of these methods often lead to efficiency issues and a lack of interpretability. Our study introduces a structural framework called SimDetector, which is a Local–Global Multi-Scale Similarity Contrast network. Specifically, the restructured and enhanced GRU module extracts more generalized local features, including long-term cyclical trends. The multi-scale sparse attention module efficiently extracts multi-scale global features with pattern information. Additionally, we modified the KL divergence to suit the characteristics of time series anomaly detection, proposing a symmetric absolute KL divergence that focuses more on overall distribution differences. The proposed method achieves results that surpass or approach the State-of-the-Art (SOTA) on multiple real-world datasets and synthetic datasets, while also significantly reducing Multiply-Accumulate Operations (MACs) and memory usage.

Robust Left-Right Hashing Scheme for Ubiquitous Computing

Ubiquitous computing systems possess the capability to collect and process data, which is subsequently shared with other devices. These systems encounter resource challenges such as memory constraints, processor speed limitations, power consumption considerations, and the availability of data storage. Therefore, maintaining data access and query processing speed in ubiquitous computing is challenging. Hashing is crucial to search operations and has caught the interest of many researchers. Several hashing techniques have been proposed and Cuckoo Hashing is found efficient to use in several applications. There are two variants of Cuckoo Hashing: Parallel Cuckoo Hashing and Sequential Cuckoo Hashing. Cuckoo Hashing suffers from challenges like high insertion latency, inefficient memory usage, and data migration. This paper proposes two hashing schemes: Left-Right Hashing and Robust Left-Right Hashing that successfully address and solve the major challenges of Sequential Cuckoo Hashing. The proposed schemes adopt the Combinatorial Hashing technique after modification and use this with a new collision resolution technique called Left-Right Random Probing. Left-Right Random Probing is a variant of random probing and uses prime numbers and Fibonacci numbers. In addition, this paper proposes a new performance indicator, degree of dexterity to estimate the performance of hashing techniques. Sequential Cuckoo Hashing suffers from hidden switching costs which are identified and its estimation is given by a new performance indicator called, T.R.C./Key. Performance of Sequential Cuckoo Hashing is order dependent.

DELTA: Memory-Efficient Training via Dynamic Fine-Grained Recomputation and Swapping

To accommodate the increasingly large-scale models within limited-capacity GPU memory, various coarse-grained techniques, such as recomputation and swapping, have been proposed to optimize memory usage. However, these methods have encountered limitations, either in terms of inefficient memory reduction or diminished training performance. In response to this, our paper introduces DELTA, an innovative approach for memory-efficient large-scale model training that combines fine-grained memory optimization and prefetching technology to reduce memory usage while maintaining high training throughput concurrently. Initially, we formulate the problem of memory-throughput joint optimization as an easy-solving 0/1 Knapsack problem. Leveraging this formalization, we use an improving polynomial complexity heuristic algorithm to address the problem effectively. Furthermore, we introduce a novel bidirectional prefetching technology into dynamic memory management, which significantly accelerates the model training when compared to relying solely on recomputation or swapping. Finally, DELTA offers users an automated training execution library, eliminating the need for manual configuration or specialized expertise. Experimental results demonstrate the effectiveness of DELTA in reducing GPU memory consumption. Compared to state-of-the-art methods, DELTA achieves substantial memory savings ranging from 40% to 72%, while maintaining comparable convergence performance for various models, including ResNet-50, ResNet-101, and BERT-Large. Notably, DELTA enables the training of GPT2-Large and GPT2-XL with batch sizes increased by 5.5 × and 6 ×, respectively, showcasing its versatility and practicality in enabling large-scale model training on GPU hardware.

Implementation of Quick Response Code Tags for Livestock Identification and Farm Management Based on Web and Android for Ruminant Farms

Technological advancements impact various sectors of life, including animal husbandry, which is an important part of the education curriculum. Innovations are needed to optimize livestock management, especially ruminant farming, which is crucial for producing meat and milk. SMKN 1 Bawen, with its study program focusing on ruminant farming (cows, sheep, and goats), still uses conventional recording methods, leading to inefficiencies and limited information for potential buyers. To address this, a QR Code-based livestock management and identification system was developed and integrated into a website and Android application using agile methods. Black box testing confirmed that all features function well. Stress testing on the website with 2 Mbps bandwidth showed 98% successful requests and 2% failed requests, with a data transmission speed of 52,816 bits per second and a reception speed of 268,624 bits per second. At 50 Mbps bandwidth, the system had a 100% success rate, with a data sending speed of 44,304 bits per second and a receiving speed of 343,272 bits per second, demonstrating good and stable performance across different bandwidths. The Android application, tested on a POCO X3 with 6 GB RAM and 128 GB internal storage, showed a highest CPU usage of 39%, memory usage of 326 MB, and network speeds of 10.3 KB/s for data received and 535.9 KB/s for data sent. These results indicate that the application runs optimally and is suitable for use.

A novel framework for developing a machine learning-based forecasting model using multi-stage sensitivity analysis to predict the energy consumption of PCM-integrated building

Accurate machine learning (ML) predictions for the early stages of the building design are crucial to construct energy-efficient buildings utilizing limited resources. Several studies have employed ML methods for energy consumption (EC) prediction without considering the utmost crucial PCM-integrated building design parameters. In addition, reducing the dataset by considering only the most significant building design parameters before applying the ML-based method would be beneficial for reducing the computational power and memory usage of the system as well as utilizing less time in the modelling process. To this end, this research presents a novel framework to establish the most robust and reliable ML-based prediction model with less complexity, considering only the most influential PCM-integrated building design parameters. These parameters were identified for future scenarios of hot semi-arid (BSh) climate zones using multi-stage sensitivity analysis. Afterward, a reduced EC database based on the most significant building's early-design-stage parameters (EDSPs) was utilized to formulate several multi-expression programming (MEP) and support vector machines (SVM)-based forecasting models, considering the variations in their hyperparameter values. Formulated prediction models have shown less time utilization through the training and testing phases for the EC evaluations of selected PCM-integrated building compared to the physical-modelling process. Several statistical parameters were used to test and validate the performance of the formulated prediction models. The acquired model evaluation and validation results demonstrated that the MEP-based prediction model (MEP15) exhibited the highest level of reliability and accuracy, showing an R2 value of >95% for both the training and testing phases. The model's interpretability showed that, throughout the parametric analysis, the developed prediction model adhered to the system's physical boundary constraints. Also, the best-performing prediction model showed energy savings of up to 12% for building integrated with PCM having a melting temperature of 28 °C. Conclusively, this research demonstrated that the developed MEP-based prediction model could be employed to precisely forecast the EC for selected PCM-incorporated building in the whole BSh climate, considering important EDSPs while utilizing minimal resources.

An archive‐based method for efficiently handling small file problems in HDFS

SummaryHadoop distributed file system (HDFS) performs well when storing and managing large files. However, its performance significantly decreases when dealing with massive small files. In response to this problem, a novel archive‐based solution is proposed. The archive refers to merging multiple small files into larger data files, which can effectively reduce the memory usage of the NameNode. The current archive‐based solutions have the disadvantages of long access time, long archive construction time, and no support for storage, updating and deleting small files in the archive system. Our method utilizes a dynamic hash function to distribute the metadata of small files across multiple metadata files. We construct a primary index that combines dynamic and static indexes for these metadata files. Regarding data files, include some read‐only files and one readable–writable file. A small file's contents are written into a readable and writable file. Upon reaching a predetermined threshold, the readable–writable file transitions into read‐only status, with a fresh readable–writable file replacing it. Experimental results show that the scheme improves the efficiency of archive access and archive creation and is more efficient than the original HDFS storage and update efficiency.

Efficacy of Transcranial Direct Current Stimulation on the Dorsolateral Prefrontal Cortex in Reducing Craving, Depression, and Improving Working Memory in Patients with Cannabis Use Disorders: A Randomized, Double-blind, Sham-controlled

Background: The beneficial effect of transcranial direct current stimulation (tDCS) on substance use disorders, achieved by reducing craving, impulsivity, and risk-taking behavior, can be considered an effective therapeutic approach. Transcranial direct current stimulation can be considered an effective therapeutic approach for reducing cannabis craving, improving levels of depression, and enhancing working memory in cannabis users. Objectives: This study investigated the effectiveness of tDCS applied to the Dorsolateral prefrontal cortex (DLPFC) on craving, depression, and working memory in cannabis users. Methods: This was a randomized, sham-controlled, double-blind, parallel-group study. Fifty male participants who met the criteria for cannabis use disorder according to DSM-5 were randomly assigned to either the real tDCS group (25 participants) or the sham tDCS group (25 participants). Pretests, including marijuana urine tests, the Marijuana Craving Questionnaire-Short Form (MCQ-SF), the Hamilton Depression Rating Scale (HDRS), and the Wechsler Memory Scale, were conducted before the intervention. The real tDCS group underwent 20 sessions of tDCS stimulation, each lasting 20 minutes, with a current intensity of 2 milliamperes applied at the F3 and F4 positions for DLPFC stimulation. The sham tDCS group received only 30 seconds of initial electrical stimulation, with no further current delivered. Both groups also received standard psychological treatments alongside the tDCS interventions. After completing the interventions, the same tests were administered again to assess the outcomes. Results: Participants who received real tDCS showed significant improvement in levels of craving, depression, and working memory. This reduction in scores and levels was statistically significant compared to the sham group. Conclusions: Transcranial direct current stimulation can be considered an effective therapeutic approach for reducing cannabis craving, improving levels of depression, and enhancing working memory in cannabis users.

Empowering IoT connectivity: unveiling the potential of IoT.Js for enhanced interconnectivity

ABSTRACT The concept of a global Internet of Things (IoT), as proposed by Bart, revolutionizes connectivity by assigning unique digital identities to everyday objects, thereby enabling the development of smarter devices. However, the advancement and expansion of the IoT face significant obstacles and constraints, including energy efficiency and memory usage, which require further investigation. To address these challenges, this study presents a novel framework named IoT.js. This framework aims to facilitate the interconnection of lightweight devices within the web-based IoT ecosystem. These devices, ranging from microcontrollers to those with limited memory capacities, stand to benefit from IoT.js’s utilization of JavaScript for IoT application development. By promoting collaboration among lightweight devices with small memory sizes, IoT.js offers a promising solution to the resource constraints encountered in IoT deployments. This study contributes to the advancement of the IoT landscape by introducing a new approach that not only enables the efficient use of limited resources but also facilitates the seamless interconnection of lightweight devices within the IoT network. Additionally, the study conducts performance analysis to measure the efficacy of the proposed framework. Through rigorous evaluation of key metrics, such as energy efficiency and memory usage, the research findings validate the effectiveness of IoT.js in addressing the challenges associated with IoT deployment. Overall, the proposed IoT.js based approach represents a significant step forward in enhancing interconnectivity within the IoT ecosystem, thereby unlocking new possibilities for the development of innovative IoT applications and services.

Simulation‐Training‐Based Deep Learning Approach to Microscopic Ghost Imaging

Herein, deep learning‐ghost imaging (DLGI) based on a digital micromirror device is realized to avoid the difficulties of a charge‐coupled device (CCD) scientific camera being unable to obtain the sample images in extremely weak illumination conditions and to solve the problem of the inverse relationship between imaging quality and imaging time in practical applications. Deep learning for computational ghost imaging typically requires the collection of a large set of labeled experimental data to train a neural network. Herein, we demonstrate that a practically usable neural network can be prepared based on the simulation results. The acquisition results of the CCD scientific camera and the simulation results with low sampling are used as the training set (1000 observations) and we can complete the data acquisition process within one hour. The results show that the proposed DLGI method can be used to significantly improve the quality of the reconstructed images when the sampling rate is 60%. This method also reduces the imaging time and the memory usage, while simultaneously improving the imaging quality. The imaging results of the proposed DLGI method have great significance for application in clinical diagnosis.

Accelerating the neural network controller embedded implementation on FPGA with novel dropout techniques for a solar inverter

Accelerating neural network (NN) controllers is important for improving the performance, efficiency, scalability, and reliability of real-time systems, particularly in resource-constrained embedded systems. This paper introduces a novel weight-dropout method for training neural network controllers in real-time closed-loop systems, aimed at accelerating the embedded implementation for solar inverters. The core idea is to eliminate small-magnitude weights during training, thereby reducing the number of necessary connections while ensuring the network’s convergence. To maintain convergence, only non-diagonal elements of the weight matrices were dropped. This dropout technique was integrated into the Levenberg–Marquardt and Forward Accumulation Through Time algorithms, resulting in more efficient training for trajectory tracking. We executed the proposed training algorithm with dropout on the AWS cloud, observing a performance increase of approximately four times compared to local execution. Furthermore, implementing the neural network controller on the Intel Cyclone V Field Programmable Gate Array (FPGA) demonstrates significant improvements in computational and resource efficiency due to the proposed dropout technique leading to sparse weight matrices. This optimization enhances the suitability of the neural network controller for embedded environments. In comparison to Sturtz et al. (2023), which dropped 11 weights, our approach eliminated 18 weights, significantly boosting resource efficiency. This resulted in a 16.40% reduction in Adaptive Logic Modules (ALMs), decreasing the count to 47,426.5. Combinational Look-Up Tables (LUTs) and dedicated logic registers saw reductions of 17.80% and 15.55%, respectively. However, the impact on block memory bits is minimal, showing only a 1% improvement, indicating that memory resources are less affected by weight dropout. In contrast, the usage of Memory 10 Kilobits (MK10s) dropped from 97 to 87, marking a 10% improvement. We also propose an adaptive dropout technique to further improve the previous results.

Efficient Numerical Implementation of the Time-Fractional Stochastic Stokes–Darcy Model

This paper presents an efficient numerical method for the fractional-order generalization of the stochastic Stokes–Darcy model, which finds application in various engineering, biomedical and environmental problems involving interaction between free fluid flow and flows in porous media. Unlike the classical model, this model allows taking into account the hereditary properties of the process under uncertainty conditions. The proposed numerical method is based on the combined use of the sparse grid stochastic collocation method, finite element/finite difference discretization, a fast numerical algorithm for computing the Caputo fractional derivative, and a cost-effective ensemble strategy. The hydraulic conductivity tensor is assumed to be uncertain in this problem, which is modeled by the reduced Karhunen–Loève expansion. The stability and convergence of the deterministic numerical method have been rigorously proved and validated by numerical tests. Utilizing the ensemble strategy allowed us to solve the deterministic problem once for all samples of the hydraulic conductivity tensor, rather than solving it separately for each sample. The use of the algorithm for computing the fractional derivatives significantly reduced both computational cost and memory usage. This study also analyzes the influence of fractional derivatives on the fluid flow process within the fractional-order Stokes–Darcy model under uncertainty conditions.

Secure multi-server coded caching

This paper considers multiple cache-enabled end-users connected to multiple servers through a linear network. We also prevent an external eavesdropper from obtaining any information about the original files in cache-aided networks. The eavesdropper does not exist in the placement phase. He can analyze and capture the multi-cast messages in the content delivery phase. He receives a single linear combination derived from each multi-cast message. Hence, we only consider the security of the delivery phase. Our work generalizes the secure caching problem with one server to a multi-server setup. A secure centralized multi-server coded caching scenario is considered, and closed-form coding delay and secret shared key storage expression are provided. Regarding our security guarantee, we show that the delivery phase does not reveal any information to the eavesdropper in terms of the mutual information metric. We analyze the system’s performance in terms of coding delay and guarantee the security of our scheme using the mutual information metric. Numerical evaluations verify that security incurs a negligible cost in terms of memory usage when the number of files and users increases, i.e., the secure and insecure bounds almost coincide. Also, we numerically show that our proposed scheme outperforms the secure coded caching problem with one server.

APPS: Authentication-enabled privacy protection scheme for secure data transfer in Internet of Things

The Internet of Things (IoT) is an emerging field that encompasses several heterogeneous devices and smart objects that are integrated with the network. In open platforms, these objects are deployed to present advanced services in numerous applications. Innumerable security-sensitive data is generated by the IoT device and therefore, the security of these devices is an important task. This work formulates a secure data transfer technique in IoT, named Authentication enabled Privacy Protection (APPS) scheme for resource-constrained IoT devices. The proposed scheme demonstrates resilience against various attacks; such as resisting reply attacks, device anonymity, untracebility, session key establishment, quantum attacks and resisting MITM attacks. For the privacy protection scheme, the secured data transfer is initiated between the entities, like IoT devices, servers, and registration centers, by using various phases, namely registration phase, key generation phase, data encryption, authentication, verification, and data retrieval phase. Here, a mathematical model is designed for protecting data privacy using hashing, encryption, secret keys, etc. Finally, performance of proposed APPS model is analyzed; wherein the outcomes reveal that the proposed APPS model attained the maximum detection rate of 0.85, minimal memory usage of 0.497MB, and minimal computational time of 112. 79 sec and minimal turnaround time 131.91 sec.

Rapid prediction for the transient dispersion of leaked airborne pollutant in urban environment based on graph neural networks

Rapidly predicting airborne pollutant dispersion in urban is vital for ventilation design and evacuation planning. Computational fluid dynamics (CFD) simulations are commonly used to provide accurate predictions, but the computational cost is too high. Although graph neural networks (GNNs) provide fast predictions of flow fields by manipulating unstructured mesh on GPU, they suffer from high memory usage and accuracy decreases when applied to large-scale urban scenes. Moreover, it is difficult for GNNs to learn the coupled relationship between wind field and pollutant concentration field. We propose a multi-objective GNN model as CFD surrogate to rapidly predict the transient dispersion of airborne pollutant under the influence of complex wind field patterns in urban environment. Based on random urban layouts generated by a 2D bin packing algorithm, we employ a validated CFD model to construct a sample dataset of wind fields and concentration fields. We leverage graph pooling and multi-scale feature fusion to improve prediction accuracy, and subgraph partitioning of both wind field and concentration field to reduce GPU memory usage. The results show that our GNN model at its best runs 1–2 orders of magnitude faster than CFD simulation with accuracy evaluation metrics R2=0.92, and achieves 70 % GPU memory reduction.

Sparse FIR Filter Design using Double Generalized Orthogonal Matching Pursuit (DGOMP)

In this paper, sparse FIR filter was designed using Double Generalized Orthogonal Matching Pursuit (DGOMP) to reduce memory usage and increasing the speed thereby decreasing computational complexity of the algorithm. Mathematical models were formulated and simulations were conducted to validate the performance of the proposed method. The performance was compared with BOMP and Conventional FIR filter. The results showed that the DGOMP method achieved higher sparsity and a better approximation of an ideal filter. Additionally, the designed sparse FIR filters using DGOMP showed better performance in terms of time of execution when the signal lengths keep increasing, giving a 10% faster execution time when compared to BOMP. The passband and stopband attenuation, as well as ripple values were better, offering the flexibility of parameter adjustment. The results showed that DGOMP is a promising approach for designing sparse FIR filters.