CPU Utilization Research Articles

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.

TAM: A Front-End to an Auto-Parallelizing Compiler

The multi-core architecture has revolutionized the parallel computing. Despite this, the modern age compilers have a long way to achieve auto-parallelization. Through this paper, we introduce a language that encouraging the auto-parallelization. We are also introducing Front-End for our auto-parallelizing compiler. Later, we examined our compiler employing a different number of core and verify results based on different metrics based on total compilation time, memory utilization, power utilization and CPU utilization. At last, we learned that parallelizing multiple files engage more CPU resources, memory and energy, but it finishes the task at hand in less time. In this paper, we have proposed a loop code generation technique that makes the generation of nested loop IR code faster by dividing the blocks into some extra code blocks using a modular approach. Our TAM compiler technique speedup by 7.506, 5.283 and 2.509 against sequential compilation when we utilized 8, 4 and 2 cores respectively. We observed that the CPU utilization of the TAM compiler reaches the maximum permissible limit when an optimal parallelizable instance is compiled.

A Blockchain and AI Based Vaccination Tracking Framework for Coronavirus (COVID-19) Epidemics

The Coronavirus disease, i.e. (COVID-19) pandemic outbreaks has inevitably lead to a corresponding movement restrictive measure by the members of both central and state government leadership. The current limitation of healthcare systems cannot predict the COVID-19 outbreaks and vaccination drive. The need of the hour is to use Blockchain Technology for the construction of immutable ledger. In this paper a Blockchain and Artificial Intelligence (AI) enabled COVID-19 vaccine tracking system has been proposed. The proposed system is designed using the Ethereum Virtual Machine (EVM), decentralized storage by Inter Planetary File System (IPFS) and implemented using Truffle and Ganache Tool. The smart contract interaction among the entities is developed with Drizzle and the front-end using ReactJS . The performance of the proposed framework has been presented with Keccack 256 transaction hash, the total number of transactions, gas consumed by each contract. Additionally the performance parameters such as latency, throughput, traffic in, out, CPU, and memory utilization of the blockchain framework are also calculated. The artificial neural network (ANN) has been used to classify the vaccination group. Such an attempt is a worthwhile addition to the state of the art as evident from the results presented herein. The proposed framework provides a way for vaccinated members of the community to be tracked and indexed so as to avail the benefits, for example, less restrictive movements nationally, internationally, or simply in their own respective communities. This paper leverages the advantages of Blockchain Technology to obtain a reliable and accurate way to facilitate the entire tracking process.

Edge-based individualized anomaly detection in large-scale distributed solar farms

Power output from large-scale solar farms is often plagued by anomalies that can adversely impact grid integration. This paper presents an anomaly detection system that used Siamese-twin neural networks for anomaly detection on edge devices in a solar farm. The model achieved an F1-score of 0.88 and was evaluated using two multi-threading schemes on a Raspberry PI, Nvidia Nano and Google Coral. A single analytics edge device could service 512 solar panels at 1 Hz. The best hardware platform was Nvidia’s Nano using a TensorFlow Lite model consuming about 35 Wh over 12 h, and with maximum CPU utilization not exceeding 60%.

X-MAN: A Non-Intrusive Power Manager for Energy-Adaptive Cloud-Native Network Functions

Emerging microservices demand flexible low-latency processing of network functions in virtualized environments, e.g., as containerized network functions (CNFs). While ensuring highly responsive low-latency CNF processing, the computing environments should conserve energy to reduce costs. In this systems integration study, we develop and evaluate the novel XDP-Monitoring Energy-Adaptive Network Functions (X-MAN) framework for managing the CPU operational states (P-states) so as to reduce the power consumption while prioritizing low-latency service. Architecturally, X-MAN consists of lightweight traffic monitors that are attached to the virtual network interfaces in the kernel space for per-CNF traffic monitoring and a power manager in user space with a global view of the CNFs on a CPU core. Algorithmically, X-MAN monitors the CPU core utilization via hybrid simple and weighted moving average prediction fed by the traffic monitors and a power management based on step-based CPU core frequency (P-state) adjustments. We evaluate X-MAN through extensive measurements in a real physical testbed operating at up to 10 Gbps. We find that X-MAN incurs significantly shorter and more consistent monitoring latencies for the CPU utilization than a state-of-the-art CPU hardware counter approach. Also, X-MAN achieves more responsive CPU core frequency adjustments and more pronounced reductions of the CPU power consumption than a state-of-the-art code instrumentation approach. We make the X-MAN source code publicly available.

The Application of Wireless Network Technology in the Capacity Building of Anhui Industry Associations Based on the Characteristics of Big Data

Recent advances of wireless networks and communication technologies, integrated into the considerable development in the field of environmental policies and public participation, have resulted in the development of wireless network-enabled big data environment in the capacity building of Anhui industry association. At the same time, effective task scheduling in the cloud server based big data environment is essential to schedule the tasks accurately and rapidly. There are thousands of jobs to be executed by the resources available on cloud data centers to achieve minimum time, high performance, and the proper utilization of CPU and resources. The necessity to fulfil user requirements is the main reason of having studies for optimizing the cloud computing of big data in terms of latency, bandwidth, execution time, and resource utilization. Therefore, this work presents a crow search optimization-based task scheduling scheme (CSO-TSS) for capacity building of Anhui industry association on wireless network-enabled big data environment. The proposed CSO-TSS technique mainly intends to schedule the tasks exist in the wireless network-enabled big data environment. Besides, the CSO-TSS technique is executed on a MapReduce environment in order to proficiently handle the big data. In addition, the CSO-TSS technique derives an objective function intending to maximize resource utilization and minimize execution time of the tasks. For examining the improved task scheduling performance of the CSO-TSS technique, a wide range of simulations were carried out and the results are investigated under several aspects. The comparative result analysis stated the better outcomes of the CSO-TSS technique over the recent approaches in terms of different measures.

Multimodal cloud resources utilization forecasting using a Bidirectional Gated Recurrent Unit predictor based on a power efficient Stacked denoising Autoencoders

To reap the advantages of many continual growing cloud services, cloud industries should adopt smart and holistic resources scheduling strategies. By deploying efficient deep learning technologies, many chaotic cloud traffics’ potential issues may be solved. Toward efficient cloud instances rightsizing and scheduling, we adopt in this paper a new Bidirectional Gated Recurrent Unit predictor based on a power efficient Stacked Denoising Autoencoders to forecast simultaneously future hourly virtual CPU, memory, and storage utilizations. Using various data ranges of resources under three AWS instances families, the best forecasting results achieved so far [1,83,30,78,331,11] of mean RMSE values and [1,37,21,63,245,13] of mean MAE values while predicting respectively future vCPU, memory, and storage utilizations. In addition, the proposed model also proved its precision stability and outperformance over the three considered SDAE-GRU, SDAE-LSTM and BiGRU benchmark models. Given the neglected power consumption measurement noticed in most related studies, we eventually validated the proposed predictor’s power efficiency by measuring in addition its real time consumed power in watt and temperature throughout the training process duration. The proposed predictor decreased the average consumed power by 5% compared to a classical sparse AE-BiGRU.

A Load-Aware Multistripe Concurrent Update Scheme in Erasure-Coded Storage System

Erasure coding has been widely deployed in today’s data centers for it can significantly reduce extra storage costs while providing high storage reliability. However, erasure coding introduced more network traffic and computational overhead in the data update process. How to improve the efficiency and mitigate the system imbalance during the update process in erasure coding is still a challenging problem. Recently, most of the existing update schemes of erasure codes only focused on the single stripe update scenario and ignored the heterogeneity of the node and network status which cannot sufficiently deal with the problems of low update efficiency and load imbalance caused by the multistripe concurrent update. To solve this problem, this paper proposes a Load-Aware Multistripe concurrent Update (LAMU) scheme in erasure-coded storage systems. Notably, LAMU introduces the Software-Defined Network (SDN) mechanism to measure the node loads and network status in real time. It selects nonduplicated nodes with better performance such as CPU utilization, remaining memory, and I/O load as the computing nodes for multiple update stripes. Then, a multiattribute decision-making method is used to schedule the network traffic generated in the update process. This mechanism can improve the transmission efficiency of update traffic and make LAMU adapt to the multistripe concurrent update scenarios in heterogeneous network environments. Finally, we designed a prototype system of multistripe concurrent updates. The extensive experimental results show that LAMU could improve the update efficiency and provide better system load-balancing performance.

A multi-layered intrusion detection system for software defined networking

The majority of existing DDoS defense mechanisms in SDN impose a significant computational burden on the controller and employ limited flow statistics and packet features. Tackling these issues, this paper presents a multi-layer defense mechanism that detects and mitigates three distinct types of flooding DDoS attacks. In the proposed framework, the detection process consists of flow-based and packet-based attack detection mechanisms employing Extreme Learning Machine-based Single-hidden Layer Feedforward Networks (ELM-SLFNs) and Case-based Information Entropy (C-IE), respectively. Moreover, the affected switches are avoided in the optimal path determined by the Floyd-Warshall algorithm, where the switches are classified based on the Hidden Markov Model (HMM) using the extracted packet features. Our simulation demonstrates the improved performance of our framework compared to similar schemes proposed in the literature in terms of different metrics, including attack detection rate, detection accuracy, false-positive rate, switch failure ratio, packet loss rate, response time, and CPU utilization.

Construction of Water Conservancy and Energy Engineering Structure Platform Based on Cloud Computing

In view of the current water conservancy engineering system in solving the problem of water conservancy spatial information sharing and repeated investment in the construction of GIS systems by various water conservancy business departments, the research on the construction of water conservancy engineering structure platform based on cloud computing is proposed. We discussed the key technologies to realize the platform: pooled services, elastic cloud services, and spatial information services. The aggregation pooling service mainly uses the pooling mechanism and improves the responsiveness of platform services by creating object instances in advance. Based on the Cloud API provided by the cloud computing platform (such as creating a virtual machine, starting a virtual machine at runtime, and obtaining the real-time status of a virtual machine) and the REST Admin API provided by the water conservancy basic service module, using cloud computing technology ideas, we designed and implemented elastic cloud service technology. Enable the water conservancy spatial information service platform to be deployed in the cloud computing environment, support cloud architecture, and provide flexible computing capabilities, because the three types of services have different requirements for CPU and network utilization, the weight coefficient can be assigned to them. The CPU slice map service is 0.1, the spatial query service is 0.3, and the spatial analysis processing service is 0.6. The network bandwidth slice map service is 0.5, the spatial query service is 0.3, and the spatial analysis processing service is 0.2. Reduce the waste of infrastructure resources; meet the concurrent access needs of a large number of users. Spatial information service aggregation is one of the core service capabilities of the platform; it can effectively aggregate multisource spatial information, generate new services, and provide new information.

Sustainable Technical Debt-Aware Computing Model for Virtual Machine Migration (TD4VM) in IaaS Cloud

In the cloud, optimal CPU and memory utilization can lead to low energy consumption, which is an important aspect of green computing. However, constantly changing workloads may contribute to resource over- or underutilization. The former violates the service level agreement’s quality of service constraints. The latter indicates that as workload decreases, virtual machine resource utilization decreases. They introduce difficult decision-making tasks when dynamically adapting (e.g., migrating) a virtual machine in order to maximize its resource utilization over time. To address these challenges, we propose a newer mathematical model called the technical debt-aware computing model for virtual machine migration (TD4VM). The model promotes a holistic approach to dynamic virtual machine adaptation for cloud service providers and addresses existing issues regarding logical aspects of virtual machine adaptation in a highly dynamic cloud environment, which includes a measurement mechanism and estimation guidelines for estimating future debt and utility. Technical debt-aware models make decisions based on VM operating costs, quality, minimizing SLA violations, and incurring technical debt. This approach connects decisions about virtual machine migration that affect overall utility over time. Our method can determine whether a virtual machine should be moved when it is over or underutilized based on its technical debt. The experimental results on a dataset obtained from the Materna-trace-1 demonstrate that the proposed approach outperforms other state-of-the-art methods on a variety of performance metrics. A numerical comparison shows that TD4VM outperforms the other approaches, with VM resource economies of 171.84%, 91.33%, 97.85%, and 93.89% for TD4VM, LRMMT, IQRMC, and IQRMMT, respectively. Additionally, we quantify the debt amassed using TD4VM and state-of-the-art techniques. When compared to LRMMT, IQRMC, and IQRMMT, which cost (in $) 0.77, 0.73, and 0.76, respectively, TD4VM accumulates the minimum debt of 0.17.

Lightweight Secure Key Distribution Protocol (LSKDP) for Wireless Sensor Networks

Because of the enormous technological revolution, the most difficult tasks that humans cannot perform are now easily addressed by small digital gadgets. An example of this technological revolution is Wireless Sensor Networks (WSN). In the WSN, numerous heterogeneous types of sensors work together to sense, monitor, capture, process, and control the physical condition of a specific region or object. The WSN has been used extensively in a number of key areas, particularly in the medical, agricultural, military, and traffic management sectors. The little electronic gadget that forms the WSN is composed of three major components: a sensing component, a power storage component, and a wireless transmission component. Despite the numerous benefits of this modern technology, security remains an unresolved concern. The goal of this work is to develop an efficient Lightweight Secure Key Distribution Protocol (LSKDP) for enhancing the process's safety in heterogeneous medical WSNs. Additionally, security and performance analyses were conducted to demonstrate the suggested methodology's efficacy. The security analysis is performed to ascertain the proposed method's vulnerability to network threats. Additionally, performance analysis is used to determine the CPU utilization, memory consumption, processing power, and computational overhead of the suggested methodology. The experimental results indicate that the suggested strategy detects attacks more accurately and with a lower computational overhead.

Efficient Cloudlet Allocation to Virtual Machine to Impact Cloud System Performance

Performance is an essential characteristic of any cloud computing system. It can be enhance through parallel computing, scheduling and load balancing. This work evaluates the connection between the response time (RT) and virtual machine’s (VM) CPU utilization when cloudlets are allocated from the datacenter broker to VM. To accentuate the RT and VM’s CPU utilization, a set of 100 and 500 heterogeneous cloudlets are analyzed under hybridized provisioning, scheduling and allocation algorithm using CloudSim simulator. These includes space shared (SS) and time shared (TS) provisioning policy, shortest job first (SJF), first come first search (FCFS), round robin (RR) and a novel length-wise allocation (LwA) algorithm. The experimental analysis shows that the RT is the least when SJF is combined with RR allocation at 40.665 secs and VM’s CPU utilization is the least when SJF is combined with LwA policy at 12.48 in all combinations of SS and TS provisioning policy.

VNGuarder: An Internal Threat Detection Approach for Virtual Network in Cloud Computing Environment

Edge-assisted Internet of things applications often need to use cloud virtual network services to transmit data. However, the internal threats such as illegal management and configuration to cloud platform intentionally or unintentionally will lead to virtual network security problems such as malicious changes of user network and hijacked data flow. It will eventually affect edge-assisted Internet of things applications. We propose a virtual network internal threat detection method called VNGuarder in a cloud computing environment, which can effectively monitor whether the virtual network configuration of legitimate users under the IaaS cloud platform has been maliciously changed or destroyed by insiders. First, based on the life cycle of cloud virtual network services, we summarized two types of internal attacks involving illegal use of virtualization management tools and illegal invocation of virtual network-related processes. Second, based on normal behavior of tenants, a hierarchical trusted call correlation scheme is proposed to provide a basis for discovering that insiders illegally call virtualized management tools and virtual network-related processes on the controller node of the cloud platform or the network node and compute node. Third, a trace-enable mechanism combining real-time monitoring and log analysis is introduced. By collecting and recording the complete call process of virtual network management and configuration in the cloud platform, and comparing it with the result of the hierarchical trusted call correlation, abnormal operations can be reported to the tenants in time. Comprehensive simulation experiments on the Openstack platform show that VNGuarder can effectively detect illegal management and configuration of virtual networks by insiders without significantly affecting the creation time of tenant networks and the utilization of CPU and memory.

Ransomware-Resilient Self-Healing XML Documents

In recent years, various platforms have witnessed an unprecedented increase in the number of ransomware attacks targeting hospitals, governments, enterprises, and end-users. The purpose of this is to maliciously encrypt documents and files on infected machines, depriving victims of access to their data, whereupon attackers would seek some sort of a ransom in return for restoring access to the legitimate owners; hence the name. This cybersecurity threat would inherently cause substantial financial losses and time wastage for affected organizations and users. A great deal of research has taken place across academia and around the industry to combat this threat and mitigate its danger. These ongoing endeavors have resulted in several detection and prevention schemas. Nonetheless, these approaches do not cover all possible risks of losing data. In this paper, we address this facet and provide an efficient solution that would ensure an efficient recovery of XML documents from ransomware attacks. This paper proposes a self-healing version-aware ransomware recovery (SH-VARR) framework for XML documents. The proposed framework is based on the novel idea of using the link concept to maintain file versions in a distributed manner while applying access-control mechanisms to protect these versions from being encrypted or deleted. The proposed SH-VARR framework is experimentally evaluated in terms of storage overhead, time requirement, CPU utilization, and memory usage. Results show that the snapshot size increases proportionately with the original size; the time required is less than 120 ms for files that are less than 1 MB in size; and the highest CPU utilization occurs when using the bzip2. Moreover, when the zip and gzip are used, the memory usage is almost fixed (around 6.8 KBs). In contrast, it increases to around 28 KBs when the bzip2 is used.

Statistical analysis of an improved tuning method for optimizing performances of Hadoop applications

Hadoop is one among upcoming and most promising technologies for handling immense volume of data, generated from a variety of sources. It supports parallel processing among multiple jobs, and when such jobs are being executed on Hadoop, it is of paramount importance to know if Hadoop cluster resources are being completely utilized or not, which in turn tells us about efficiency in resource utilization for the purpose of effective completion of jobs. It could happen that Hadoop cluster resources are not being completely utilized, and because of which there arises a need for performance tuning. Fortunately, Hadoop framework equips us with several parameters to do the same and fine-tune aforesaid jobs. Performance tuning comprises of four main components: CPU utilization, Memory consumption, Disk I/O and Network traffic. In the proposed work, authors have presented several important relative parameters in relation to aforesaid components, along with how to fine-tune these parameters with appropriate values so as to optimize Hadoop execution.

Run-Time Adaptive In-Kernel BPF/XDP Solution for 5G UPF

Flexibility is considered a key feature of 5G softwarization to deliver a timely response to changes in network requirements that may be caused by traffic variation, user mobility, dynamic network function chains, slice lifecycle management operations, among others. In this article, we evolve the upf-bpf1 open-source project by proposing a new design to improve its flexibility by reducing the run-time adaptation time. The project proposes an in-kernel solution based on BPF and eXpress Data Path (XDP) for 5G User Plane Function (UPF) implementations. The Just-In-Time (JIT) compilation may have a huge impact on the adaptation time due to the in-kernel verification of the BPF programs at run-time. Our results show latency improvements of around 95% to inject the BPF program into the Linux kernel. Furthermore, the solution keeps the same functionalities and delivers a packet processing performance of around 10–11 Mpps using 6 cores with almost 70% of the CPU utilization in downlink/uplink directions.

GiantVM: A Novel Distributed Hypervisor for Resource Aggregation with DSM-aware Optimizations

We present GiantVM, 1 an open-source distributed hypervisor that provides the many-to-one virtualization to aggregate resources from multiple physical machines. We propose techniques to enable distributed CPU and I/O virtualization and distributed shared memory (DSM) to achieve memory aggregation. GiantVM is implemented based on the state-of-the-art type-II hypervisor QEMU-KVM, and it can currently host conventional OSes such as Linux. (1) We identify the performance bottleneck of GiantVM to be DSM, through a top-down performance analysis. Although GiantVM offers great opportunities for CPU-intensive applications to enjoy the aggregated CPU resources, memory-intensive applications could suffer from cross-node page sharing, which requires frequent DSM involvement and leads to performance collapse. We design the guest-level thread scheduler, DaS (DSM-aware Scheduler), to overcome the bottleneck. When benchmarking with NAS Parallel Benchmarks, the DaS could achieve a performance boost of up to 3.5×, compared to the default Linux kernel scheduler. (2) While evaluating DaS, we observe the advantage of GiantVM as a resource reallocation facility. Thanks to the SSI abstraction of GiantVM, migration could be done by guest-level scheduling. DSM allows standby pages in the migration destination, which need not be transferred through the network. The saved network bandwidth is 68% on average, compared to VM live migration. Resource reallocation with GiantVM increases the overall CPU utilization by 14.3% in a co-location experiment.

The Network Model of Internet Access Intranet Based on Embedded Platform

<p>To better manage and control the embedded devices located in the remote Intranet, the functional requirements of the Internet accessing the Intranet need to be effectively solved. This paper studies and implements a network model of external network accessing intranet network and realizes the network model’s function by programming. At the same time, an experimental platform is built to test the performance metrics of the network model, such as the maximum load and CPU utilization, device connection success rate, user access response time, and other performance parameters. The test results show that the network model can meet the network scale of at least 3000 intranet devices and has high practical value and application prospects.</p> <p> </p>

Ascertain the efficient machine learning approach to detect different ARP attacks

Software-Defined Networking (SDN) is a programmable network architecture that allows network devices to be controlled remotely, but it is still highly susceptible to traditional attacks such as Address Resolution Protocol (ARP) Poisoning, ARP Flooding, and others. The classification of benign network traffic from ARP Poison and ARP Flooding attacks is presented in this paper employing machine learning (ML) techniques. A python application is developed at the SDN controller using Mininet that collects and logs the features required to detect the attack into a file known as a traffic dataset. This dataset is used to train the ML model and detect the attacks. The hybrid model of Convolution Neural Network-Long Short Term Memory (CNN-LSTM) model out-performs the other ML models with an accuracy score of 99.73%. During the attack, a high CPU utilization of more than 97% and a high memory usage serve as experimental evidence. The attack detection time of 63000 microseconds also demonstrates the efficiency of attack detection.