Server Usage Research Articles

Deep learning for the PSIPRED Protein Analysis Workbench.

The PSIRED Workbench is a long established and popular bioinformatics web service offering a wide range of machine learning based analyses for characterizing protein structure and function. In this paper we provide an update of the recent additions and developments to the webserver, with a focus on new Deep Learning based methods. We briefly discuss some trends in server usage since the publication of AlphaFold2 and we give an overview of some upcoming developments for the service. The PSIPRED Workbench is available at http://bioinf.cs.ucl.ac.uk/psipred.

Intelligent Resource Orchestration for 5G Edge Infrastructures

The adoption of edge infrastructure in 5G environments stands out as a transformative technology aimed at meeting the increasing demands of latency-sensitive and data-intensive applications. This research paper presents a comprehensive study on the intelligent orchestration of 5G edge computing infrastructures. The proposed Smart 5G Edge-Cloud Management Architecture, built upon an OpenNebula foundation, incorporates a ONEedge5G experimental component, which offers intelligent workload forecasting and infrastructure orchestration and automation capabilities, for optimal allocation of virtual resources across diverse edge locations. The research evaluated different forecasting models, based both on traditional statistical techniques and machine learning techniques, comparing their accuracy in CPU usage prediction for a dataset of virtual machines (VMs). Additionally, an integer linear programming formulation was proposed to solve the optimization problem of mapping VMs to physical servers in distributed edge infrastructure. Different optimization criteria such as minimizing server usage, load balancing, and reducing latency violations were considered, along with mapping constraints. Comprehensive tests and experiments were conducted to evaluate the efficacy of the proposed architecture.

Ethereum Smart Contract Deployment for a Real Estate Management System (REMS) Implemented in Blockchain

The real industry is crucial sector of the world wide financial system, with significant economic, social and environmental implications. However, real estate transactions are often slow, complex and costly and can be prone to fraud errors, which can lead to significant financial losses and legal disputes. This paper proposes the adoption of blockchain technology in real estate cadastral systems as a solution to the challenges encountered in managing property ownership and transactions. We have described the implementation of our prototype called REMS (Real Estate Management System) and we have presented the deployment of smart contracts in Ethereum platform. Based on measurement, benchmarks and other observation of the system, we have evaluated the server usage of the blockchain network and decided whether Blockchain-as-a-Service (BaaS) should be involved or not in our system. The study results demonstrate the successful implementation of a real estate management system (REMS) using blockchain technology and Ethereum's smart contracts. This study is important because it confirms that similar solutions can be implemented in other areas of public administration, where the structure of the work is similar, i.e. where we deal with issuing documents to citizens.

A Q-Learning-Based Load Balancing Method for Real-Time Task Processing in Edge-Cloud Networks

Edge computing has emerged as a promising solution to reduce communication delays and enhance the performance of real-time applications. However, due to the limited processing power of edge servers, it is challenging to achieve efficient task processing. In this paper, we propose a Q-learning-based load-balancing method that optimizes the distribution of real-time tasks between edge servers and cloud servers to reduce processing time. The proposed method is dynamic and adaptive, taking into account the constantly changing network status and server usage. To evaluate the effectiveness of the proposed load-balancing method, extensive simulations are conducted in an Edge-Cloud network environment. The simulation results demonstrate that the proposed method significantly reduces processing time compared to traditional static load-balancing methods. The Q-learning algorithm enables the load-balancing system to dynamically learn the optimal decision-making strategy to allocate tasks to the most appropriate server. Overall, the proposed Q-learning-based load-balancing method provides a dynamic and efficient solution to balance the workload between edge servers and cloud servers. The proposed method effectively achieves real-time task processing in edge computing environments and can contribute to the development of high-performance edge computing systems.

An energy-efficient topology-aware virtual machine placement in Cloud Datacenters: A multi-objective discrete JAYA optimization

By the advent of cloud computing and the numerous related web applications, data center networks (DCNs) are becoming complex to provide all-to-all communications between underlying devices; it is done by spending a huge amount of electricity consumption. Energy consumption management is the first class concern for cloud providers in this energy hungry devices and also for the green computing goals. In the large DCs, the power management of hundreds or even many thousands idle switches can be a promising approach toward overall cost reduction. The virtual machine placement (VMP) scheme which is aware of both VMs affinity and underlying network topology for co-hosting dependent VMs as physically near as possible and lowering down power state of idle devices can enhance sustainability objectives. On the other hand, resource wastage lowers system utilization which leads more physical server usage causing more power consumption as a consequence. To address the issue, this paper presents an energy-efficient topology-aware VM placement scheme in the cloud DCs which is formulated to a multi-objective optimization problem with power consumption and resource wastage minimization perspective. To deal with this combinatorial problem, an advanced multi-objective discrete version of JAYA (MOD-JAYA) algorithm is presented since the search space of VMP is discrete in nature. The proposed algorithm is validated by intensively variable circumstances and simulations upon conducted scenarios. The simulation results prove the superiority of the proposed advanced MOD-JAYA algorithm in solving VMP problem in comparison with other existing schemes in terms of prominent assessment metrics.

QUEUEING SYSTEM IN FAIR PRICE SHOP OF KERALA TO MITIGATE WAITING TIME AND OPTIMIZATION OF SERVERS

Queues are a typical occurrence in everyday life, particularly during peak hours in Fair Price Shops (FPS), government offices, banks, post offices, super markets, hospitals, and clinics. The arrival rate, service rate, and server usage all play a role in the queueing process of any system. We expect a balance between waiting time and server utilization in an effective system, with the ideal being less waiting time and maximum server utilization. This paper investigates efficient queue management in FPS using M/M/C queueing model. The major goal of this research is to minimize customer waiting time in the FPS with the minimal number of servers. It demonstrates the application of the simulation modelling as a method for optimizing the number of servers in a specific FPS in Kerala, India. The model was developed to provide the outcomes of system parameters like arrival rate, waiting time and server utilization.

Performance Framework for Virtual Machine Migration in Cloud Computing

In the cloud environment, the transfer of data from one cloud server to another cloud server is called migration. Data can be delivered in various ways, from one data centre to another. This research aims to increase the migration performance of the virtual machine (VM) in the cloud environment. VMs allow cloud customers to store essential data and resources. However, server usage has grown dramatically due to the virtualization of computer systems, resulting in higher data centre power consumption, storage needs, and operating expenses. Multiple VMs on one data centre manage share resources like central processing unit (CPU) cache, network bandwidth, memory, and application bandwidth. In multi-cloud, VM migration addresses the performance degradation due to cloud server configuration, unbalanced traffic load, resource load management, and fault situations during data transfer. VM migration speed is influenced by the size of the VM, the dirty rate of the running application, and the latency of migration iterations. As a result, evaluating VM migration performance while considering all of these factors becomes a difficult task. The main effort of this research is to assess migration problems on performance. The simulation results in Matlab show that if the VM size grows, the migration time of VMs and the downtime can be impacted by three orders of magnitude. The dirty page rate decreases, the migration time and the downtime grow, and the latency time decreases as network bandwidth increases during the migration time and post-migration overhead calculation when the VM transfer is completed. All the simulated cases of VMs migration were performed in a fuzzy inference system with performance graphs.

LOCUS: User-Perceived Delay-Aware Service Placement and User Allocation in MEC Environment

In the multi-access edge computing environment, app vendors deploy their services and applications at the network edges, and edge users offload their computation tasks to edge servers. We study the user-perceived delay-aware service placement and user-allocation problem in edge environment. We model the MEC-enabled network, where the user-perceived delay consists of computing delay and transmission delay. The total cost in the offloading system is defined as the sum of service placement, edge server usage and energy consumption cost, and we need to minimize the total cost by determining the overall service-placing decision and user-allocation decision, while guaranteeing that the user-perceived delay requirement of each user is fulfilled. Our considered problem is formulated as a Mixed Integer Linear Programming problem, and we prove its NP-hardness. Due to the intractability of the considered problem, we propose a LOCal-search based algorithm for USer-perceived delay-aware service placement and user-allocation in edge environment, named LOCUS, which starts with a feasible solution and then repeatedly reduces the total cost by performing local-search steps. After that, we analyze the time complexity of LOCUS and prove that it achieves provable guaranteed performance. Finally, we compare LOCUS with other existing methods and show its good performance through experiments.

ReAD: Reliability aware data

As transistor technology scales beyond 32 nm, the reliability of the processors and servers reduces at an exponential rate. In recent times, the qualification of reliability of servers has reduced from 5 years to 3 years. With this current trend, the lifetime of servers can be expected to shrink further. In a cloud data center, with hundreds and hundreds of servers, the ill effect of shorter lifetime reliability will be magnified. Reliability of servers influences the performance, cost and power efficiency of a cloud data center. In this paper, a cost-effective technique to continuously monitor the reliability of servers and scheduling tasks based on these reliability values is proposed. The proposed technique monitors the hardware parameters for the current executing workload to calculate the reliability metrics. The reliability metric used in ReAD is MTTF (Mean Time to Fail) and this is calculated for individual servers. The load balancer reads the MTTF of individual servers and performs the load balance, considering reliability as one of the parameters. The proposed technique is non-invasive in nature and does not require the server to go into a special test mode. This saves the testing time. The experimental analysis shows on an average 25% improvement in Server Usage Metric with the minimum improvement around 5% and the maximum is around 50%.

Makespan Optimisation in Cloudlet Scheduling with Improved DQN Algorithm in Cloud Computing

Despite increased cloud service providers following advanced cloud infrastructure management, substantial execution time is lost due to minimal server usage. Given the importance of reducing total execution time (makespan) for cloud service providers (as a vital metric) during sustaining Quality-of-Service (QoS), this study established an enhanced scheduling algorithm for minimal cloudlet scheduling (CS) makespan with the deep Q-network (DQN) algorithm under MCS-DQN. A novel reward function was recommended to enhance the DQN model convergence. Additionally, an open-source simulator (CloudSim) was employed to assess the suggested work performance. Resultantly, the recommended MCS-DQN scheduler revealed optimal outcomes to minimise the makespan metric and other counterparts (task waiting period, resource usage of virtual machines, and the extent of incongruence against the algorithms).

New developments including X-ray standing waves in the dynamical Bragg diffraction program of X-ray Server.

X-ray Server (https://x-server.gmca.aps.anl.gov) is a collection of programs for online modelling of X-ray diffraction and scattering. The dynamical diffraction program is the second most popular Server program, contributing 34% of total Server usage. It models dynamical X-ray diffraction from strained crystals and multilayers for any Bragg-case geometry including grazing incidence and exit. This paper reports on a revision of equations used by the program, which yields ten times faster calculations in most use cases, on implementing calculations of X-ray standing waves and on adding new options for modelling diffraction from monolayers.

Fuzzy Request Set Modelling for Detecting Multiplexed Asymmetric DDoS Attacks on HTTP/2 servers

The introduction of HTTP/2 has led to a dramatic change in web traffic. The steady flow of requests in HTTP/1.1 has been replaced by bursts of multiple requests, largely due to the introduction of multiplexing in HTTP/2 which allows users to send multiple requests through a single connection. This feature was introduced in order to reduce the page loading time by multiplexing a web page and its associated resources in a single connection. While this feature has significantly improved user experience, it can be misused to launch sophisticated application layer DDoS attacks against HTTP/2 servers. Instead of the intended use of multiplexing, attackers can force the web server to process multiple random requests simultaneously, leading to increased server usage. The use of computationally intensive requests can further exacerbate the situation. These attacks, called Multiplexed Asymmetric Attacks, pose a dangerous threat to HTTP/2 servers and stem from the lack of verification of the multiplexed requests. In this work, an approach to model an HTTP/2 request set as a fuzzy multiset is presented. The proposed approach uses a combination of relative cardinality and request workload to detect multiplexed AL-DDoS attacks. Experiments on open source datasets demonstrate that the proposed approach is able to detect multiplexed AL-DDoS attacks with an accuracy of around 95%, while maintaining a low False Positive Rate (FPR) of around 3%.

Hybrid LRU Algorithm for Enterprise Data Hub using Serverless Architecture

In hybrid LRU algorithm was built to execute parameterized priority queue using Least Recently Used model. It helped to determine the object in an optimum mode to remove from cache. Experiment results demonstrated ~30% decrease of the execution time to extract data from cache store during object cache extraction process. In the era of modern utility computing theory, Serverless architecture is the cloud platform concept to hide the server usage from the development community and runs the code on-demand basis. This paper provides Hybrid LRU algorithm by leveraging Serverless Architecture benefits. Eventually, this new technique added few advantages like infrastructure instance scalability, no server management, reduced cost on efficient usage, etc. This paper depicts about the experimental advantage of Hybrid LRU execution time optimization using Serverless architecture.

Integrated Data Bank Development as Supporting System of Civil Service Management in Middle Vocational School

The objectives of this study are: 1) Improved Integrated Data Bank services for staffing data and information; 2) Enhanced support system for data and information service server usage; 3) Improving data and information services that are accurate for Staffing Management. The research method used is Research and Development (R&D). Data collection techniques using questionnaires and interviews. Product trials were conducted twice, the first trial involved 15 people, the second trial involved 35 people. The results of the study are 1) Integrated Data Bank services for employee data and information can be used effectively and efficiently. This is based on system reliability with a success rate of 100%. In addition, the efficiency test using GTMetrix gives the results of the PageSpeed Grade B (85%) with a loading rate of the system page 7.4 seconds; 2) the support system for the use of data and information service servers runs according to the Integrated Data Bank application; 3) Personnel management data and information services can be used as an accurate staffing database and show that the Integrated Data Bank staffing management information system can be used as a supporting management system for staff at SMK Muhammadiyah Karangpucung; 4) Integrated Data Bank Products have gone through the process of design validation, instrument validation, and product validation through judgment experts and experts, which produce data in the form of Data Flow Diagrams (DFD), Unified Modeling Language (UML) design, and Entity Relationship Diagram Bank designs Integrated Data. Based on the results of the product design validity test showed that the layout (layout), interface (interface), and content (content) of the system itself obtained a final percentage of 84%. Suggestions in this study include 1) Socialization of the Integrated Data Bank system to employees about the existence of this system can be optimized so that all information can be conveyed properly; 2) Integrated Data Bank systems can be utilized optimally to integrate staffing data; 3) Integrated Data Bank System can be developed into a superior product as a staffing database system in the SMK environment.

Towards Secure, Flexible and Efficient Role Based Hospital’s Cloud Management System: Case Study

INTRODUCTION: Many organizations of health care have recognized that quality of service can be improved by maintaining e-records of patient’s reports, medical histories, surgeries recordings, etc. over the multimedia cloud servers. But, the data breach is always a matter of stake for an organization as well as for the patients.OBJECTIVES: This work has considered a qualitative scenario related to multimedia e-content management for a multiforte hospital’s cloud server.METHODS: An End-to-End Encryption with the cryptographic algorithm is applied along with an access control framework for secure transmission and storage of e-records.RESULTS: The results are presented in the form of bar graphs for the time taken to perform encryption/decryption of various media files and combo graphs for different scenarios of access control.CONCLUSION: This can provide prevention from many attacks/threats, ensures authentication and privacy along with limiting the server usage and cost.

Queue Systems Evaluation of Fast food Industrial

The queuing system can influent customer satisfaction and always interesting to analyze. The queuing at HokBen use a Single Channel Multi-Phase system, 1 service to offer and prepare order menus and 1 cashier. The study was conducted at the Hokben Ujung Menteng Store in East Jakarta, for one week, at 4 sessions during rush hour. After the data obtained, a Single Channel Multi-Phase queue parameter was calculated, namely: the average arrival per unit time, the average service unity of time, the level of server usage, the number of customers in the queue, the number of customers in the system, the waiting time in the queue, the waiting time in the system, and customer probabilities in the system. The result showed that the customer's arrival pattern meets the Poison distribution and the service pattern meets the Normal distribution, but the service time still greater than the standard service time, and of course not satisfy to customers, because they waited a bit longer. After a multi-channel multiphase simulation, which used 2 service lines, the service time smaller than the standard service time, and service more efficient. This research becomes useful for Hokben management specifically, and also for the fast-food industry in general.
 
 Keywords: Queue, Single channel single-phase, multi-channel multi-phase

A Method of Fault Fix Priority Identification for Open Source Project

Open source software are adopted as embedded systems, server usage because of quick delivery, cost reduction and standardization of systems. Many open source software are developed under the peculiar development style known as bazaar method. According to this method, faults are detected and fixed by developers around the world, and the fixed result will be reflected in the next release. Also, the fix time of faults tends to be shorter as the development of open source software progresses. However, several large-scale open source projects have a problem that faults fixing takes a lot of time because the faults corrector cannot handle many faults reports quickly. In this paper, we aim to identify the fix priority of newly registered faults in the bug tracking system by using random forest, and we make an index to detect the faults that require high fix priority and long fault fixing time when faults are reported in specific version of open source project. The index is derived and identified by using open source project data obtained from bug tracking system. In addition, we try to improve the detection accuracy of the proposed index by learning not only the specific version but also the fault report data of the past version by using random forest considering the characteristic similarities of faults fix among different versions. As a result, the detection accuracy has highly improved comparing with using only specific version data and using logistic regression.

Prediction of Fault Fix Time Transition in Large-Scale Open Source Project Data

Open source software (OSS) programs are adopted as embedded systems regarding their server usage, due to their quick delivery, cost reduction, and standardization of systems. Many OSS programs are developed using the peculiar style known as the bazaar method, in which faults are detected and fixed by developers around the world, and the result is then reflected in the next release. Furthermore, the fix time of faults tends to be shorter as the development of the OSS progresses. However, several large-scale open source projects encounter the problem that fault fixing takes much time because the fault corrector cannot handle many fault reports. Therefore, OSS users and project managers need to know the stability degree of open source projects by determining the fault fix time. In this paper, we predict the transition of the fix time in large-scale open source projects. To make the prediction, we use the software reliability growth model based on the Wiener process considering that the fault fix time in open source projects changes depending on various factors such as the fault reporting time and the assignees to fix the faults. In addition, we discuss the assumption that fault fix time data depend on the prediction of the transition in fault fixing time.

Dynamic Resource Provisioning on Fog Landscapes

A huge amount of smart devices which have capacity of computing, storage, and communication to each other brings forth fog computing paradigm. Fog computing is a model in which the system tries to push data processing from cloud servers to “near” IoT devices in order to reduce latency time. The execution orderings and the deployed places of services make significant effect on the overall response time of an application. Beside new research directions in fog computing, e.g., fog-cloud collaboration, service scalability, fog scalability, mobile fog computing, fog federation, trade-off between energy consumption and communication efficiency, duration of storing data locally, storage security and communication security, and semantic-aware fog computing, the service deployment problem is one of the attractive research fields of fog computing. The service deployment is a multiobjective optimization problem; there are so many proposed solutions for various targets, such as response time, communication cost, and energy consumption. In this paper, we focus on the optimization problem which minimizes the overall response time of an application with awareness of network usage and server usage. Then, we have conducted experiments on two service deployment strategies, called cloudy and foggy strategies. We analyze numerically the overall response time, network usage, and server usage of those two strategies in order to prove the effectiveness of our proposed foggy service deployment strategy.

Deploying Chains of Virtual Network Functions: On the Relation Between Link and Server Usage

Recently, network function virtualization has been proposed to transform from network hardware appliances to software middleboxes. Normally, a demand needs to invoke several virtual network functions (VNFs) following the order determined by the service chain along a routing path. In this paper, we study the joint problem of the VNF placement and path selection to better utilize the network. We discover that the relation between the link and server usage plays a crucial role in the problem. Inspired by stress testing, we first propose a systematic way to elastically tune the link and server usage of each demand based on the network status and properties of demands. In particular, we compute a proper routing path length, and decide, for each VNF in the service chain, whether to use additional server resources or to reuse resources provided by existing servers. We then propose a chain deployment algorithm that follows the guidance of this link and server usage. Via simulations, we show that our design effectively adapts resource allocation to network dynamics and, hence, serves more demands than other heuristics.