Real-time Cloud Research Articles

Joint Information Freshness and Completion Time Optimization for Vehicular Networks

The demand for real-time cloud applications has seen an unprecedented growth over the past decade. These applications require rapidly data transfer and fast computations. This article considers a scenario where multiple IoT devices update information on the cloud, and request a computation from the cloud at certain times. The time required to complete the request for computation includes the time to wait for computation to start on busy virtual machines, performing the computation, waiting and service in the networking stage for delivering the output to the end user. In this context, the freshness of the information is an important concern and is different from the completion time. This article proposes novel scheduling strategies for both computation and networking stages. Based on these strategies, the age-of-information (AoI) metric and the completion time are characterized. A convex combination of the two metrics is optimized over the scheduling parameters. The problem is shown to be convex and thus can be solved optimally. Moreover, based on the offline policy, an online algorithm for job scheduling is developed. Numerical results demonstrate significant improvement as compared to the considered baselines.

Knowledge Transfer Using User-Generated Data within Real-Time Cloud Services

When automatic speech recognition (ASR) is provided as a cloud service, it is easy to collect voice and application domain data from users. Harnessing these data will facilitate the provision of more personalized services. In this paper, we demonstrate our transfer learning-based knowledge service that built with the user-generated data collected through our novel system that deliveries personalized ASR service. First, we discuss the motivation, challenges, and prospects of building up such a knowledge-based service-oriented system. Second, we present a Quadruple Transfer Learning (QTL) method that can learn a classification model from a source domain and transfer it to a target domain. Third, we provide an overview architecture of our novel system that collects voice data from mobile users, labels the data via crowdsourcing, utilises these collected user-generated data to train different machine learning models, and delivers the personalised real-time cloud services. Finally, we use the E-Book data collected from our system to train classification models and apply them in the smart TV domain, and the experimental results show that our QTL method is effective in two classification tasks, which confirms that the knowledge transfer provides a value-added service for the upper-layer mobile applications in different domains.

An automatic lane identification method for the roadside light detection and ranging sensor

The roadside Light Detection and Ranging (LiDAR) sensor can provide the high-resolution micro traffic data (HRMTD) of all road users by collecting real-time point clouds in three-dimensional (3D) space. The HRMTD collected by the roadside LiDAR provides a solution to fill the data gap under the mixed situation (both connected vehicles and unconnected vehicles exist on the roads) for connected vehicle technologies. Lane identification is important information in HRMTD. The current lane identification algorithms are mainly developed for autonomous vehicles, which could not be directly used to process roadside LiDAR data. This article provides an innovative algorithm to automatically identify traffic lanes for the roadside LiDAR data. The proposed lane identification algorithm includes five major steps: background filtering, point clustering, object classification, frame aggregation, and traversal search. The parameters used in the algorithm are selected by balancing the time cost and the accuracy. With the GPS information, the location of the lane can be transferred into the GoogleEarth and be compared with the location of the lane in real world. The testing results showed that the average distance error (ADE) compared to the real location in Google Earth was less than 0.1 m. This robust lane identification can release engineers from the manual lane identification task and avoid any error caused by manual work. The extracted lane locations can be used for researchers and practitioners to locate the vehicles precisely in different applications.

An Energy and Performance Aware Scheduler for Real-Time Tasks in Cloud Datacentres

Datacentres provide the foundations for cloud computing, but require large amounts of electricity for their operation. Approaches that promise to reduce power use by minimizing execution time, for example using different scheduling and resource management techniques, are discussed in the literature. This paper summarizes some of the most important scheduling techniques in clouds focusing on power consumption, covering VM-level, host-level and task-level scheduling where the most promising approach is task level scheduling, with energy savings by means of load filtering, consolidation, adapted CPU throughput, or host power control. We explore use of the rate monotonic (RM) and backfilling algorithms for real-time task scheduling in cloud environment because RM is the simplest fixed priority scheduling technique, and thus the choice for modern real-time systems, and prior uses of RM in task scheduling have demonstrated power efficiency with optimal results. We specifically consider deadline-based tasks scheduling for real-time clouds which, to the best of our knowledge, has not been employed previously. RM with backfilling is experimentally evaluated and results show that, compared to the classical algorithms, all tasks were scheduled with minimum power consumption (5.5% - 29.3%), on minimum resources (3.9% - 25.2% less) while majority were meeting their deadlines (93.21% - 94.7%). The approach can guarantee deadline oriented Software as a Service (SaaS) in cloud if arrival rate i.e. network transfer time can be estimated in advance. We subsequently provided an extension of the proposed approach to task-based load balancing for almost balanced resource utilization and approximately 1.0% to 1.6% energy efficiency.

Real-Time Cloud Visual Simultaneous Localization and Mapping for Indoor Service Robots

Unlike traditional industrial robots, indoor service robots are usually required to possess high intelligence, such as the skills of flexible moving, precise spacial perceiving. And high intelligence is always accompanied by consuming complicated and expensive computation resources. One solution for indoor service robots is centralization of expensive computation resource so that it is possible to design a low cost client with a high-intelligence brain. However, as a fundamental intelligence function for mobile indoor robots, if a real-time visual Simultaneously Localization and Mapping (vSLAM) system is split into client and brain, it will be confronted with new challenges, such as the barrier of instant data sharing and performance degradation brought by network delay in between. To solve the problem, we focus on a framework and approach of cloud-based visual SLAM in this paper, and provide an efficient solution to offload the expensive computation and reduce the cost of robot clients. The integrated system is distributed in a 3-level Cloud with light-weight tracking, high precision dense mapping, and map sharing. Based on recent excellent algorithms, our system is able to run a real-time sparse tracking on the client, and a real-time dense mapping on the cloud server, which outputs an explicit 3D dense map. Only keyframes are sent to the local cloud center to reduce the network bandwidth requirement. Dense geometric pose estimation besides feature-based methods is computed to make the system resistant to feature-less indoor scenes. The camera poses associated with keyframes are optimized on the local computing cloud center, and are sent back to the client to decrease the trajectory drift. We evaluate the system on the Technical University of Munich (TUM) datasets, Imperial College London and National University of Ireland Maynooth (ICL-NUIM) datasets, and the real data captured by our robot in terms of visual odometry on the client side and dense maps generated on the server cloud. Qualitative and quantitative experiments show our cloud visual SLAM system is able to bear the network delay in Local Area Network (LAN), and it is an efficient vSLAM solution for indoor service robots with high intelligence from a centric brain.

Research on Financial Efficiency of Yunnan Baiyao Based on Big Data

The analysis of corporate financial efficiency can meet the different decision needs of users of accounting information. Big data transforms accounting information from real-time data to real-time cloud computing platforms, providing a rich information foundation for financial efficiency analysis of enterprises. This article uses the financial data of Yunnan Baiyao in the past six years, and analyzes the problems and countermeasures of its financial management by analyzing the company’s operating capacity, profitability, solvency and development ability, in order to provide useful reference for enterprise management.

Reconstructing 3D point clouds in real time with look-up tables for structured light scanning along both horizontal and vertical directions.

By scanning static, not moving, objects along both the horizontal and vertical axes instead of one, structured light illumination achieves more accurate and robust 3D surface reconstructions but with greater latency on computing 3D point clouds. If scanning is performed along only one axis, it has been reported that look-up tables, manually derived from the calibration matrices of a camera and a projector, can significantly help to speed up computation; however, it has been nearly impossible to manually derive similar look-up tables for phases scanned along two axes. In this Letter, we bridge this divide by introducing the constraint of epipolar geometry to automatically compute look-up tables and thus, significantly speed up computing 3D point clouds with only basic arithmetic operations rather than time-consuming matrix computations. Experimental results show that the proposed method, using only single-thread CPU computing, reduces process latency by an order of magnitude.

Pose Prediction of Autonomous Full Tracked Vehicle Based on 3D Sensor.

Autonomous vehicles can obtain real-time road information using 3D sensors. With road information, vehicles avoid obstacles through real-time path planning to improve their safety and stability. However, most of the research on driverless vehicles have been carried out on urban even driveways, with little consideration of uneven terrain. For an autonomous full tracked vehicle (FTV), the uneven terrain has a great impact on the stability and safety. In this paper, we proposed a method to predict the pose of the FTV based on accurate road elevation information obtained by 3D sensors. If we could predict the pose of the FTV traveling on uneven terrain, we would not only control the active suspension system but also change the driving trajectory to improve the safety and stability. In the first, 3D laser scanners were used to get real-time cloud data points of the terrain for extracting the elevation information of the terrain. Inertial measurement units (IMUs) and GPS are essential to get accurate attitude angle and position information. Then, the dynamics model of the FTV was established to calculate the vehicle’s pose. Finally, the Kalman filter was used to improve the accuracy of the predicted pose. Compared to the traditional method of driverless vehicles, the proposed approach was more suitable for autonomous FTV. The real-world experimental result demonstrated the accuracy and effectiveness of our approach.

Intercomparisons of Cloud Mask Products Among Fengyun-4A, Himawari-8, and MODIS

In this paper, we developed a unified and operational cloud mask algorithm for the new-generation geostationary (GEO) meteorological satellite imagers of the Advanced Geostationary Radiation Imager (AGRI) aboard Fengyun-4A (FY-4A) and the Advanced Himawari Imager (AHI) aboard Himawari-8 (H08). We investigated the all-round performance of the cloud mask algorithm. Spatiotemporally, the algorithm matches the official Collection-6 cloud mask products of a Moderate Resolution Imaging Spectroradiometer (MODIS) from both the Terra and Aqua platforms, which we employed as the benchmark for performing intercomparisons and validations. The robust cloud mask algorithm can show high consistency between FY-4A/AGRI and H08/AHI. The MODIS-based validation results suggest that cloudy scene identification is better than that observed for clear skies for both FY-4A/AGRI and H08/AHI; there is also a relatively low false-alarm ratio (FAR). Moreover, the algorithm is more reliable during daytime hours, with a hit rate (HR) of approximately 92% for both FY-4A/AGRI and H08/AHI. We found slightly higher accuracy in cloud-masking results over water than those over land. Furthermore, we found that more than 67% of the matched pixels for both advanced GEO imagers had no bias when taking MODIS as the benchmark. Overall, HR values were approximately 91.04% and 91.82% for FY-4A/AGRI and H08/AHI, respectively. These results confirm the high quality of the algorithm for retrieving real-time cloud mask products.

Reliability of Fault Tolerance in Cloud using Machine Learning Algorithm

The basic fault tolerance issues seen in cloud computing are identification and recovery. To fight with these issues, so many fault tolerance methods have been designed to decrease the faults. However, due to the reliability and web based service giving behavior, fault tolerance in cloud computing will be a huge challenge. The present model is not just on tolerating faults but also to decrease the possibility of future faults as well[4].The fault tolerance deals with the exact and constant operation of the fault segments. The processing on computing nodes can be done remotely in the real time cloud applications, so there could be more possibilities of errors. Hence there lies an immense necessity for fault tolerance to attain consistency to the real time computing on cloud infrastructure. The “fault tolerance” can be explained through fault processing that have two basic stages. The stages are (i) The effective error processing stage which is used to intended for carrying the “effective error” back to inactive state, i.e., before the error occurred (ii) The latent error processing stage intended for guaranteeing that the fault does not get effective once again.

An Improved Stream Processing Access

Migration of Legacy applications into modern Cloud, IOT architecture are challenging tasks and many researchers are showing interest to build modern Real time cloud and IOT based applications like smart cities, Video mining, Health care, Industrial event monitoring and many more for modern human life. Such applications should require efficient online data streaming techniques to process large amount of unstructured online data streams instead of offline. Modern customer centric applications with different verticals are looking for distributed and horizontal data streaming approaches. Many real time streaming approaches are emerging to utilize or process large real-time data by replacing legacy centralized scenarios which are causing more memory utilization, delay and fault tolerance. In this paper we present common models and architectures for real time utilization of cloud and IoT based application stream processing. Utilization of the real-time data of IoT/Cloud applications are possible with collective streaming techniques of network, data processing. In this paper we are focusing on improving stream processing techniques, limitations and future research directions for real-time stream processing

Calibration of Multi-LIDAR Systems: Application to Bucket Wheel Reclaimers

This paper presents a method to calibrate the extrinsic parameters of light detection and rangings (LIDARs) mounted on a bucket-wheel reclaimer (BWR). BWR’s are widely used for stacking and reclaiming bulk materials in stockyards. Current BWR systems are either manually operated, remotely operated, or semiautomated using the real-time point cloud data generated by one or more LIDARs. The accurate calibration is of crucial importance as the accuracy of the earth frame point cloud data depends on it. The automated calibration is also a prerequisite for fully autonomous BWR control. The calibration of BWR systems is more difficult than many other LIDAR systems because of their limited pose variation capabilities and the environmental constraints of stockyards. This paper analyzes the problem, presents observability conditions, presents a method to estimate the calibration parameters, and discusses the challenges involved with BWR systems. The results demonstrate subdecimeter accuracy.

Agent-based cloud service negotiation architecture using similarity grouping approach

Challenges and issues in the field of cloud service negotiation framework optimization have been an active area of research. During service level agreement, the probability of negotiation conflict between the service consumers and providers is high. This may arise due to aggressive behavior, selfish misperception, vague preferences and uncertain goals of the negotiating participants. One of the key challenges identified in negotiation framework is optimizing the negotiation conflict among the negotiators. In order to minimize such conflicts, existing frameworks group the negotiation pairs that contain similar and non-aggressive behavioral patterns by exploiting the distance, binary, context dependent and fuzzy similarity approaches. These approaches get better success rate only if the dimensionality of negotiator attributes is low. As emerging real-time cloud service negotiation applications are characterized by negotiation attributes of high dimensionality, the existing approaches are inappropriate for these applications. In addition, the existing approaches group the negotiation pairs using distances based measure in two-dimensional negotiation attribute, whose value will vary for high-dimensional attributes. In this work, an Angle-based Similarity Grouping (ASG) approach is proposed that appropriately groups the highly cooperative negotiation pairs and thereby increases the success rate and decreases communication overhead.

Packet Encryption for Securing Real-Time Mobile Cloud Applications

The evolution of smart devices has brought the most confidential data to mobile applications and cloud computing with its processing power is serving the huge processing requirements of these applications. However, these advancements has brought in serious data security concerns, as there is constant shuttling of data from devices to cloud and backforth, and encryption is the most commonly used technique for securing data in transmission. Every encryption technique is breakable, but its efficiency is calculated with the time it can withstand an attack. But in today’s world, as cloud computing comes with almost unlimited computing resources, the present day encryption techniques might become inadequate for securing confidential data in transmission and after transmission. Existing encryption techniques, encrypt the whole data packets in a session using the same session key. If a third party can decrypt a single packet during or after transmission, all the packets in that session can be compromised. This work proposes a packet encryption scheme, where a packet key once used will never be used throughout the session, with minimal latency and maximum security for real-time mobile cloud applications.

A Container-Based Technique to Improve Virtual Machine Migration in Cloud Computing

Hypervisor-enabled virtualization is a promising technology to deploy applications and implement Virtual Machine (VM) migration on cloud data centers. However, one critical aspect in hypervisor-based solution is that to deploy applications each VM is installed with a dedicated guest OS along with the requirement of binaries and library files and substantially large VM's image size. Therefore, it is a significant concern for data center administrators because it causes performance overheads. To solve these issues, a new paradigm of virtualization technology that has attracted considerable attention is containerization. Although, progress has been made to address container-based virtualization, there has been less attention to explore the issue of performance evaluation of container and VM in a migration environment. The main contribution of this research study is to develop an experimental setup to implement and evaluate the performance of LXD/CR container-based migration technique and compared it with the pre-copy VM migration scheme. To conduct this performance evaluation, first we modified Linux kernel and implemented LXD which uses liblxc to launch container, then migration of the running container is facilitated using a checkpoint/restore mechanism of CRIU technique. To ensure the applicability and validity for real-time cloud infrastructure, we executed a wide variety of benchmarks such as web server, CPU-intensive, disk I/O and database. From the results obtained, it is found that compared to pre-copy VM migration scheme, LXD/CR container migration technique reduces the downtime, migration time, amount of data transferred and the number of pages transferred by 75.66%, 65.55%, 76.63%, and 76.78%, respectively. In addition, the migration overhead caused is reduced by 55.89% w.r.t. CPU utilization and 76.52% w.r.t. RAM utilization. This being an important contribution for migrating the running applications efficiently and consequently increase the business value of the cloud infrastructure providers.

Task Failure Prediction using Combine Bagging Ensemble (CBE) Classification in Cloud Workflow

Scientific applications adopt cloud environment for executing its workflows as tasks. When a task fails, dependency nature of the workflows affects the overall performance of the execution. An efficient failure prediction mechanism is needed to execute the workflow efficiently. This paper proposes a failure prediction method which is implemented using various machine learning classifiers. Among different classifiers, Naive Bayes predicts the failure with the highest accuracy of 94.4%. Further, to improve the accuracy of prediction, a novel ensemble method called combine bagging ensemble is introduced and acquires overall accuracy as 95.8%. The validation of proposed method is carried out by comparing simulation and real-time cloud testbed.

Parallel Algorithms for Real-Time GPGPU Volumetric Cloud Dynamics and Morphing

Parallel Algorithms for Real-Time GPGPU Volumetric Cloud Dynamics and Morphing

IAssistMe - Adaptable Assistant for Persons with Eye Disabilities

Visually challenged people may experience certain difficulties in their daily interaction with technology. That is essentially because the main way to exchange and process information is by written text, images or videos. Since the basic purpose of innovation is to improve people’s lifestyle, in this paper we propose a system that can make technology accessible to a broader group. Our prototype is presented as a mobile application based on vocal interaction, which can help people facing visual disorders consult their personal agenda, create an event, invite other friends to attend it, check the weather in certain areas and many other day-to-day tasks. Regarding the implementation, the project consists of a mobile application that interacts with a cloud based system, which makes it reliable and low in latency due to the resource availability in multiple global regions, provided by the newly emerging platform used in building the infrastructure. The novelty of the system lays in the highly flexible serverless architecture [1] that is open to extension and closed to modification through the set of autonomous cloud processing methods that sustain the base of the functionality. This distributed processing approach guarantees that the user always receives a response from his personal assistant, either by using artificial intelligence context generated phrases, by real-time cloud function processing or by fallback to the training answers.

Optimal Data Center Scheduling for Quality of Service Management in Sensor-Cloud

The proposed work concentrates on the networking facets of sensor-cloud infrastructures—one of the first attempts of its kind. In a sensor-cloud, multiple sets of physical sensor nodes that are activated based on an application demand, in turn give rise to multiple distinct virtual sensors (VSs). The VSs are considered to span across multiple geographical regions; thereby, depositing the data (from each of the VS) to the closest cloud data center (DC). Quite obviously, multiple geospatial DCs get involve with an application. However, the principle of sensor-cloud is to store and conglomerate the data from various VSs, before they can be provisioned as Sensors-as-a-Service (Se-aaS). The assortation of data occurs within a single Virtual Machine (VM) (or in some cases multiple VMs) residing inside a particular DC. This work addresses the problem of scheduling a particular DC that congregates data from various VSs, and transmit the same to the end-user application. The work follows the general pairwise choice framework of the Optimal Decision Rule. The scheduling of the DC is performed under several network constraints, such as data migration cost, data delivery cost, and service delay of an application that ensures the preservation of the Quality-of-Service (QoS) and maintenance of the user satisfaction. The work quantifies the effective QoS of Se-aaS and determines an optimal decision rule for electing a particular DC. While arriving at a collective decision, the work incorporates the fallible decision making ability of a DC; thereby, excluding the loss of generality. Experimental results depict that the proposed algorithm for generating the optimal decision rule finds applicability in real-time cloud computing scenarios.

Energy-Efficient Adaptive Resource Management for Real-Time Vehicular Cloud Services

Providing real-time cloud services to Vehicular Clients (VCs) must cope with delay and delay-jitter issues. Fog computing is an emerging paradigm that aims at distributing small-size self-powered data centers (e.g., Fog nodes) between remote Clouds and VCs, in order to deliver data-dissemination real-time services to the connected VCs. Motivated by these considerations, in this paper, we propose and test an energy-efficient adaptive resource scheduler for Networked Fog Centers (NetFCs). They operate at the edge of the vehicular network and are connected to the served VCs through Infrastructure-to-Vehicular (I2V) TCP/IP-based single-hop mobile links. The goal is to exploit the locally measured states of the TCP/IP connections, in order to maximize the overall communication-plus-computing energy efficiency, while meeting the application-induced hard QoS requirements on the minimum transmission rates, maximum delays and delay-jitters. The resulting energy-efficient scheduler jointly performs: (i) admission control of the input traffic to be processed by the NetFCs; (ii) minimum-energy dispatching of the admitted traffic; (iii) adaptive reconfiguration and consolidation of the Virtual Machines (VMs) hosted by the NetFCs; and, (iv) adaptive control of the traffic injected into the TCP/IP mobile connections. The salient features of the proposed scheduler are that: (i) it is adaptive and admits distributed and scalable implementation; and, (ii) it is capable to provide hard QoS guarantees, in terms of minimum/maximum instantaneous rates of the traffic delivered to the vehicular clients, instantaneous rate-jitters and total processing delays. Actual performance of the proposed scheduler in the presence of: (i) client mobility; (ii) wireless fading; and, (iii) reconfiguration and consolidation costs of the underlying NetFCs, is numerically tested and compared against the corresponding ones of some state-of-the-art schedulers, under both synthetically generated and measured real-world workload traces.