Memory Utilization Research Articles

The case for distributed shared-memory databases with RDMA-enabled memory disaggregation

Memory disaggregation (MD) allows for scalable and elastic data center design by separating compute (CPU) from memory. With MD, compute and memory are no longer coupled into the same server box. Instead, they are connected to each other via ultra-fast networking such as RDMA. MD can bring many advantages, e.g., higher memory utilization, better independent scaling (of compute and memory), and lower cost of ownership. This paper makes the case that MD can fuel the next wave of innovation on database systems. We observe that MD revives the great debate of "shared what" in the database community. We envision that distributed shared-memory databases (DSM-DB, for short) - that have not received much attention before - can be promising in the future with MD. We present a list of challenges and opportunities that can inspire next steps in system design making the case for DSM-DB.

Age-Related Differences of Mandarin Tone and Consonant Aspiration Perception in Babble Noise.

This study investigated the categorical perception of Mandarin tones and consonant aspiration contrasts in babble noise among adults and adolescents aged 12-14 years, and explored the association between working memory and categorical perception. Twenty-four adults and 20 adolescents with Mandarin as their native language were recruited. Their performances of phonemic identification and discrimination in babble noise and quiet conditions, digit span tasks, and nonword repetition were assessed. Results indicated that, firstly, in the noise condition, both adults and adolescents showed wider boundary widths and lower between-category accuracies when perceiving aspiration of consonants than in the quiet condition, and the categorical boundary of tone perception in adolescents showed a transitional tendency toward Tone 1. Secondly, discrimination of consonant aspiration in adolescents needed to be further developed. Lastly, the accuracy of nonword repetition in adolescents was lower than that in adults, and adults with better auditory verbal working memory had better performance on tone perception. Our results provided evidence that tone perception is acquired easier than consonant aspiration perception, and tone perception is more robust and less susceptible to noise interference. Categorical perception performance relates to the capacity and utilization of auditory verbal working memory in some ways.

A Hybrid Supervised Learning Approach for Detection and Mitigation of Job Failure with Virtual Machines in Distributed Environments

Distributed data processing techniques are very popular nowadays due to high data generation from various resources. To increase work learning outcomes and to reduce consumption, modern massive computational systems divide jobs into several smaller tasks that perform in parallel. Nevertheless, responding with straggler processes, which are sluggish running processes that rise the total response time, is a typical performance issue in such platforms. In this paper, we proposed the detection of struggler nodes in a large distributed environment using a hybrid machine learning technique. Initially, the data has been collected from numerous virtual machine network logs. The entire data set has various fields such as Virtual Machine ID, CPU load, memory load, bandwidth utilization, etc. Memory utilization an input to the proposed system is collected from the garbage collection log files where the memory consumption on each VM and its timestamp is recorded. This is the most efficient way to get the memory consumption in web/desktop applications. Similarly, the CPU, I/O and bandwidth utilization is grabbed from the process monitoring functionality and SAR (System Activity Report) utility from the respective VM boxes. This data set is useful to identify weather-specific virtual machine is heated up or not. In this approach, we proposed three conventional machine learning algorithms and a hybrid machine learning algorithm for the identification of node status. Main purpose of the proposed system is to identify the slow performing node in an efficient way to prevent the other nodes from failures. This can provide effective load balancing and low response time for task execution from available VM’s in distributed cloud environments. To create its training program, several extractions of features approaches were used. TF-IDF, correlational co-occurrence, and density-based features have been mined from the whole data set. With extensive experimental analysis, we evaluate our system with our proposed classification algorithm. As a result, the system produces higher classification accuracy of 94.5% over the traditional machine learning classifiers. If the proposed system is tested against the data set fields, memory load and CPU load on the homogenous machine configurations, we see more efficiency while detecting the underperforming node than the heterogenous machine configurations.

Performance Test and Improvement of Computer Network Virtualization Software in Cloud Computing Environment

In order to make up for the problems of inconveniences, unavailability, differences, and exclusivity in the performance test of the computer virtual network software in the cloud computing environment, a method of building a distributed system performance test platform based on cloud computing is proposed. The distributed system performance test method based on the decision tree, the common performance index test method, and the resource scheduling method in the cloud environment is mainly adopted, and the cloud computing and the automatic performance test technology are combined so as to build a prototype system of the cloud performance test platform. Experimental data show that the CPU utilization and the memory utilization can be reduced to 32% and 30%, respectively, by using this method. It is about 30 times better than the result before application. The results show that it can improve the efficiency, availability, and convenience of software performance test greatly. Also, it also can improve the software performance to facilitate people’s life and work in the cloud computing environment.

University Educational Administration Management Platform Integrating Distributed Real-time Cloud Computing System

The prosperity of the economy has made the society develop rapidly, and the traditional education management system can no longer meet the needs of the modern society. Under the background of college enrollment expansion, it has brought new challenges to school management and development. In recent years, the evolution and widespread adoption of virtualization, service-oriented architecture, autonomous, and utility computing have converged to give birth to a new paradigm: cloud computing. Cloud computing allows on-demand delivery of software, hardware, and data as a service. Cloud computing has the characteristics of massive processing and storage capacity, high efficiency, virtualization, and low cost. It has changed the traditional way of accessing software and hardware, providing users with convenient services and computing resources, which has achieved remarkable results in the related fields. Meanwhile, cloud computing plays an important role in education and teaching through its wide application in various fields. Based on this, this paper aims to study the application of distributed real-time cloud computing system in educational administration management in colleges and universities. This paper establishes a university educational administration management platform based on the cloud service computing system, and applies the management technology and distributed storage technology of the cloud computing platform. The system constructed in this paper significantly improves the utilization of CPU and memory. The research results of this paper showed that the response speed of the educational administration management system built through cloud computing has been significantly improved, with an average response time of 1.3s and an error of less than 0.4.

Cloud-based data security transactions employing blowfish and spotted hyena optimisation algorithm

Because of its on-demand servicing and scalability features in cloud computing, security and confidentiality have converted to key concerns. Maintaining transaction information on third-party servers carries significant dangers so that malicious individuals trying for illegal access to information data security architecture. This research proposes a security-aware information transfer in thecloud-based on the blowfish algorithm (BFA) to address the issue. The user is verified initially with the identification and separate the imported data using pattern matching technique. Further, BFA is utilised to encrypt and save the data in cloud. This can safeguard the data and streamline the proof so that client cannot retrieve the information without identification which makes the environment secure. The suggested approach's performance is evaluated using several metrics, including encryption time, decryption time, memory utilisation, and runtime. Compared to the existing methodology, the investigational findings clearly show that the method takes the least time to data encryption.

In-depth analysis of dynamic degree load balancing technique in public cloud for heterogeneous cloudlets

<span lang="EN-US">Load <span lang="EN-US">balancing is one of the challenges of the distributed computing worldview. With the enormous development in clients and their interest for different administrations on the distributed computing stage, compelling or productive asset usage in the cloud climate has turned into an urgent concern. Load balancing is critical to keeping cloud computing running smoothly. This study examines the research using four scheduling algorithms: dynamic degree balance CPU based (D2B_CPU), dynamic degree balanced membership based (D2B_Membership), dynamic degree memory balanced allocation (D2MBA) and hybrid dynamic degree balance (HDDB) algorithm. Central processing unit (CPU) utilisation, bandwidth utilisation, and memory utilisation are used as performance measures to verify the performance of these algorithms. The CloudSim simulation programme was used to simulate these algorithms. The primary goal of this work is to aid in the future construction of new algorithms by researching the behaviour of various existing algorithms.</span></span>

Green Based Software Engineering Approach for Sustainable Protocol

Abstract: A sturdy green software model is the prominent necessity for various governments fields and banking areas for Energy utilization processing. The model should give all the Energy consumption aspects namely power consumption, utilization of Memory, (Co2) Carbon dioxide release which is related to development of software and make it energy efficient. For Software Development Life Cycle phases this research paper proposes a new green software model for the virtual host machine, which is the base for all the Green based Software Development process.

Enhancing the Performance of Android

Abstract: In the present day, with the increasing popularity of the android operating system, applications being built on the Android SDK draw much attention. However, some applications built on android tend to have a cumbersome interface causing inconveniences to users. This can be reduced by eliminating the redundancies during the build of these applications. List view is commonly used in android applications to arrange a list of scrollable items in a sequential order. This consumes a considerable amount of memory and CPU utilization resulting in the degradation of the applications. One proposed solution to this can be the introduction of the RecyclerView. The RecyclerView traditionally replaces the GridView and the ListView which were previously used in android applications. This improvement is achieved by recycling the views which are out of the visibility of the user inturn enhancing both the ListView as well as the GridView. Networking in android was previously a nightmare for developers where they had to retrieve and analyze their own data. Data transfer is performed in Android applications using a Web REST API. Development speed can be enhanced by efficient data analysis by utilizing a library from the entire range of REST client libraries. REST (Representational State Transfer) has become the most commonly used way for creating, publishing, and consuming Web services, exploiting JavaScript Object Notation (JSON) as a data exchange format. The most used libraries for accessing the REST Web APIs today are Android Volley and Retrofit. An android device is hindered by limited power, memory and resources. Several architectures such as MVC(Model-View-Controller), MVP(Model-view-Presenter) and MVVM(ModelView-Viewmodel) are commonly used in the development of android applications. In order to develop applications with efficient CPU utilization, reduced memory usage and a lowered execution time, the most suitable of these architectures is to be utilized

Web Personalized Recommendation Model Using Temporal Fuzzy Association Rule Mining

Abstract: A web access log file contains timely sequenced log entries which include essential fields to indicate user activities. Analysis of these patterns provides valuable information for web designer to quickly respond to their individual needs. Many industries are struggling to retain regular interested customers for the improvement of customer relationship. Retrieval of relevant information automatically from these log files for interested group of users is a difficult process, since acquiring interested user profiles which evolves continuously with respect to time are not so easy. The paper presents a novel Web Personalized Recommendation Model (WPRM) using temporal fuzzy association rule mining technique. Temporal Fuzzy Association Rule Mining (FTARM) technique is proposed and applied on a focused set of interested users to provide intelligent recommendations. The proposed model results in less execution time and reduced memory utilization with high accuracy.

Api Service Infrastructure Using Kubernetes And Terraform Based On Microservices Ngoorder.Id

The growing number of ngorder.id service users causes traffic to Api Ngorder to be higher, so a new infrastructure is needed in order to maintain Api Ngorder uptime during high traffic and also maintain service stability. In the implementation process, Api scripts that are currently running on a monolith cluster will be divided into several categories and will be split into several kubernetes clusters. To support autoscale, a Horizontal Pod Autoscaler was added, and to route traffic it would use the Api Gateway from Amazon Web Service. In this infrastructure test, it is done by testing the logic script function using Katalon Studio and testing at the infrastructure level by doing a crash test in the form of failing to deploy and terminating the pod, as well as performing a stress test to test autoscaling in the cluster, the entire test can be run by performing a stress test on the php service pods. by setting the autoscaler parameter Memory Utilization Percentage 125%, 150% and 250%, proving that the HorizontalPodAutoscaler (HPA) as an autoscaler handler can function according to the targets and parameters that have been determined.

Damage detection of grotto murals based on lightweight neural network

Non-contact damage detection of grotto murals is a challenging task, due to their minor cracks and subtle defects. In this paper, a lightweight neural network-based grotto mural damage detection algorithm, named Ghost-C3SE YOLOv5, is proposed, which can effectively reduce the network size while obtaining reasonable damage detection results. First, based on the YOLOv5 detection algorithm, the dimension of the convolution layer is reduced, by adjusting the network structure and integrating a Ghost module. Second, the channel attention mechanism is introduced into the feature extraction backbone network, to adjust the weight of each feature according to its importance, which can accelerate the convergence speed of the loss function during the model training process. Finally, the experiment results have shown that the lightweight model Ghost-C3SE YOLOv5, applied on the Pascal Voc dataset, reduces the number of parameters by about 22.55 million while ensuring the detection precision and recall rate. Also, the training time and the model size are reduced by 36.21% and 46.04%, respectively. Precision is increased by 1.29% and recall remains comparative, while the utilization rate of the GPU is improved by 41.55%. This has addressed the shortcomings of laborious process and low accuracy in manual detection of the grotto murals. Furthermore, a real-time detection performance of 44.86 FPS is achieved here for damage detection of murals in the YunGang Grottoes, which has overcome the drawbacks of high model complexity, high computational cost, slow detection speed and low memory utilization of various existing deep learning algorithms.

Fuzzy-Dedup: A secure deduplication model using cosine based Fuzzy interference system in cloud application

Storage consumption is increasing significantly these days, with consumers trying to find an effective approach to safe storage space. In these situations, a deduplication in cloud storage services is a significant way to reduce bandwidth and service space by omitting unnecessary information and keeping only a single copy of the information. This raises computational, privacy and storage issues when large numbers of handlers outsource the similar data to cloud service storage. To overcome these problems, an effective Fuzzy-Dedup framework is designed in this research by integrating four steps namely is introduced, which breaks down the data into fixed size chunks and is immediately fingerprinted by a hashing algorithm for ensuring data authentication and then indexing is done with the help of traditional b-tree indexing, similarity function is calculated to compute the similarity value in the documents. After calculating the similar values, the fuzzy interference system is designed by formulating appropriate rules for the decision-making process that determines duplicate and non-duplicate files by obtaining an effective de-duplication ratio over existing methods. After detecting duplicate files, the inline based deduplication policy checks that the new data is ready to send for storage against existing data and does not store any redundant data it discovers. The proposed model is implemented in MATLAB software is carried out several performance metrics and these parameter attained better performance such as, deduplication ratio of 1.2, memory utilization of 12500 bytes in inline and 9550 bytes in offline, throughput of 32500 Mb/s in inline and 25500 Mb/s in offline and processing time of 0.4494 s in inline and 0.1139 s in offline. Thus when compared to previous methods, such as Two Thresholds Two Divisors deduplication (TTTD) approach proposed design shows high range of performance.

Nicotine promotes the utility of short-term memory during visual search in macaque monkeys.

The central cholinergic system is a major therapeutic target for restoring cognitive functions. Although manipulation of cholinergic signaling is known to alter working memory (WM), the underlying mechanism remains unclear. It is widely accepted that WM consists of multiple functional modules, one storing short-term memory and the other manipulating and utilizing it. A recently developed visual search task and a relevant model can be used to assess multiple components of WM during administration of acetylcholine receptor (AChR)-related substances. The effects of systemic administration of AChR-related agents on WM and eye movements were examined during the oculomotor foraging task. Three monkeys performing the task received an intramuscular injection of saline or the following AChR-related agents: nicotine (24 or 56μg/kg), mecamylamine (nicotinic AChR antagonist, 1.0mg/kg), oxotremorine (muscarinic AChR agonist, 3.0µg/kg), and scopolamine (muscarinic AChR antagonist, 20μg/kg). The task was to find a target among 15 identical objects by making eye movements within 6s. The data were analyzed according to the foraging model that incorporated three parameters. Nicotine and mecamylamine significantly increased the utility but not the capacity of short-term memory, while muscarinic AChR-related agents did not alter any WM parameters. Further regression analyses with a mixed-effect model showed that the beneficial effect of nicotine on memory utility remained after considering eye movement variability, but the beneficial effect of mecamylamine disappeared. Nicotine improves visual search, mainly by increasing the utility of short-term memory, with minimal changes in oculomotor parameters.

Breaking Cuckoo Hash: Black Box Attacks

Introduced less than twenty years ago, cuckoo hashing has a number of attractive features like a constant worst case number of memory accesses for queries and close to full memory utilization. Cuckoo hashing has been widely adopted to perform exact matching of an incoming key with a set of stored (key, value) pairs in both software and hardware implementations. This widespread adoption makes it important to consider the security of cuckoo hashing. Most hash based data structures can be attacked by generating collisions that reduce their performance. In fact, for cuckoo hashing collisions could lead to insertion failures which in some systems would lead to a system failure. For example, if cuckoo hashing is used to perform Ethernet lookup and a given MAC address cannot be added to the cuckoo hash, the switch would not be able to correctly forward frames to that address. Previous works have shown that this can be done when the attacker knows the hash functions used in the implementation. However, in many cases the attacker would not have that information and would only have access to the cuckoo hash operations to perform insertions, removals or queries. This article considers the security of a cuckoo hash to an attacker that has only a black box access to it. The analysis shows that by carefully performing user operations on the cuckoo hash, the attacker can force insertion failures with a small set of elements. The proposed attack has been implemented and tested for different configurations to demonstrate its feasibility. The fact that cuckoo hash can be broken with only access to its user functions should be taken into account when implementing it in critical systems. The article also discusses potential approaches to mitigate this vulnerability.

An overview of computer programming languages: A fundamental piece of knowledge for humanity

The major goal of this review is to standardise the programming languages that are useful for controlling the behaviour of machines on computers. Syntax and semantic norms are defined in programming languages, just as they are in natural languages. Over a hundred thousand programming languages exist, with new ones appearing every year. Few languages ever gain widespread adoption, yet talented programmers may use dozens of languages over the course of their careers. The authors looked at the most prominent standized programming languages' runtime performance, memory utilisation, applications, and key pros and drawbacks in this review.

Human Behavior Recognition in Outdoor Sports Based on the Local Error Model and Convolutional Neural Network.

With the rapid development of the Internet, various electronic products based on computer vision play an increasingly important role in people's daily lives. As one of the important topics of computer vision, human action recognition has become the main research hotspot in this field in recent years. The human motion recognition algorithm based on the convolutional neural network can realize the automatic extraction and learning of human motion features and achieve good classification performance. However, deep convolutional neural networks usually have a large number of layers, a large number of parameters, and a large memory footprint, while embedded wearable devices have limited memory space. Based on the traditional cross-entropy error-based training mode, the parameters of all hidden layers must be kept in memory and cannot be released until the end of forward and reverse error propagation. As a result, the memory used to store the parameters of the hidden layer cannot be released and reused, and the memory utilization efficiency is low, which leads to the backhaul locking problem, limiting the deployment and execution of deep convolutional neural networks on wearable sensor devices. Based on this, this topic designs a local error convolutional neural network model for human motion recognition tasks. Compared with the traditional global error, the local error constructed in this paper can train the convolutional neural network layer by layer, and the parameters of each layer can be trained independently according to the local error and does not depend on the gradient propagation of adjacent upper and lower layers. As a result, the memory used to store all hidden layer parameters can be released in advance without waiting for the end of forward and backward propagation, avoiding the problem of backhaul locking, and improving the memory utilization of convolutional neural networks deployed on embedded wearable devices.

Performance study of the Memory Utilization of an Improved Pattern Matching Algorithm using Bit-Parallelism

The strategy of packing several data values in a single computer word and refreshing them all in a solitary operation is referred to bit parallelism. It assumes a significant part in pattern matching because it can handle in parallel the length of pattern sizes. In this paper, an Improved Pattern Matching model (IPM) proposed, which makes searching process quicker and decreases how much memory used in processing input data. C# was used for the development of the model. With a computer word size of 64bits and pattern length ranging from 8 characters to 72 characters, the system decides how much memory is used. The developed model was evaluated and contrasted with the existing model using 64bits computer word size (cws) and the pattern length of 72 characters. The assessment showed that the IPM had minimal worth of MU contrasted with the existing model (BNDM, SBNDM, and FSBNDM). This IPM model can be embraced for improvement of the size of string data stored in computer word because of its capacity to diminish memory space usage. w:LsdException Locked="false" Priority="51" Name="Grid Table 6 Colorful

A Multi-Dimensional and Multi-Factor Trust Computation Framework for Cloud Services

In this paper, we propose a novel trust computation framework (TCF) for cloud services. Trust is computed by taking into consideration multi-dimensional quality of service (QoS) evidence and user feedback. Feedback provides ample evidence regarding the quality of experience (QoE) of cloud service users. However, in some cases, users may behave maliciously and report false feedback. Users can carry out collusion and Sybil attacks to slander/self-promote cloud services. Trust computed in such cases could be misleading and inaccurate. Evaluating the credibility of user feedback can help in not only preventing the collusion and Sybil attacks but also remunerating the affected cloud services. Despite the advantages of credibility evaluation, very few studies take into consideration feedback credibility and multi-dimensional evaluation criteria. Considering the limitations of existing studies, we propose a new TCF in which trust is computed by aggregating multi-dimensional evidence from QoS and QoE. We have used multi-dimensional QoS attributes to compute the objective trust of cloud services. The QoS attributes are divided into three dimensions, i.e., node profile, average resource consumption, and performance. The node profile of a cloud service is attributed to CPU frequency, memory size, and hard disk capacity. The average resource consumption is quantified based on the current CPU utilisation rate, current memory utilisation rate, current hard disk utilisation rate, and energy consumption. Moreover, the performance of a cloud service is measured by the average response time and task success ratio. Besides that, the credibility of feedback is evaluated to prevent the malicious behaviour of cloud users. Our results demonstrate the effectiveness of our proposed TCF in computing accurate trust in cloud services.

Federated Learning-Inspired Technique for Attack Classification in IoT Networks

More than 10-billion physical items are being linked to the internet to conduct activities more independently and with less human involvement owing to the Internet of Things (IoT) technology. IoT networks are considered a source of identifiable data for vicious attackers to carry out criminal actions using automated processes. Machine learning (ML)-assisted methods for IoT security have gained much attention in recent years. However, the ML-training procedure incorporates large data which is transferable to the central server since data are created continually by IoT devices at the edge. In other words, conventional ML relies on a single server to store all of its data, which makes it a less desirable option for domains concerned about user privacy. The Federated Learning (FL)-based anomaly detection technique, which utilizes decentralized on-device data to identify IoT network intrusions, represents the proposed solution to the aforementioned problem. By exchanging updated weights with the centralized FL-server, the data are kept on local IoT devices while federating training cycles over GRUs (Gated Recurrent Units) models. The ensemble module of the technique assesses updates from several sources for improving the accuracy of the global ML technique. Experiments have shown that the proposed method surpasses the state-of-the-art techniques in protecting user data by registering enhanced performance measures of Statistical Analysis, Energy Efficiency, Memory Utilization, Attack Classification, and Client Accuracy Analysis for the identification of attacks.