Jing Bi

Temporal Task Scheduling With Constrained Service Delay for Profit Maximization in Hybrid Clouds

As cloud computing is becoming growingly popular, consumers’ tasks around the world arrive in cloud data centers. A private cloud provider aims to achieve profit maximization by intelligently scheduling tasks while guaranteeing the service delay bound of delay-tolerant tasks. However, the aperiodicity of arrival tasks brings a challenging problem of how to dynamically schedule all arrival tasks given the fact that the capacity of a private cloud provider is limited. Previous works usually provide an admission control to intelligently refuse some of arrival tasks. Nevertheless, this will decrease the throughput of a private cloud, and cause revenue loss. This paper studies the problem of how to maximize the profit of a private cloud in hybrid clouds while guaranteeing the service delay bound of delay-tolerant tasks. We propose a profit maximization algorithm (PMA) to discover the temporal variation of prices in hybrid clouds. The temporal task scheduling provided by PMA can dynamically schedule all arrival tasks to execute in private and public clouds. The sub problem in each iteration of PMA is solved by the proposed hybrid heuristic optimization algorithm, simulated annealing particle swarm optimization (SAPSO). Besides, SAPSO is compared with existing baseline algorithms. Extensive simulation experiments demonstrate that the proposed method can greatly increase the throughput and the profit of a private cloud while guaranteeing the service delay bound.

TTSA: An Effective Scheduling Approach for Delay Bounded Tasks in Hybrid Clouds

The economy of scale provided by cloud attracts a growing number of organizations and industrial companies to deploy their applications in cloud data centers (CDCs) and to provide services to users around the world. The uncertainty of arriving tasks makes it a big challenge for private CDC to cost-effectively schedule delay bounded tasks without exceeding their delay bounds. Unlike previous studies, this paper takes into account the cost minimization problem for private CDC in hybrid clouds, where the energy price of private CDC and execution price of public clouds both show the temporal diversity. Then, this paper proposes a temporal task scheduling algorithm (TTSA) to effectively dispatch all arriving tasks to private CDC and public clouds. In each iteration of TTSA, the cost minimization problem is modeled as a mixed integer linear program and solved by a hybrid simulated-annealing particle-swarm-optimization. The experimental results demonstrate that compared with the existing methods, the optimal or suboptimal scheduling strategy produced by TTSA can efficiently increase the throughput and reduce the cost of private CDC while meeting the delay bounds of all the tasks.

Application-Aware Dynamic Fine-Grained Resource Provisioning in a Virtualized Cloud Data Center

A key factor of win–win cloud economy is how to trade off between the application performance from customers and the profit of cloud providers. Current researches on cloud resource allocation do not sufficiently address the issues of minimizing energy cost and maximizing revenue for various applications running in virtualized cloud data centers (VCDCs). This paper presents a new approach to optimize the profit of VCDC based on the service-level agreements (SLAs) between service providers and customers. A precise model of the external and internal request arrival rates is proposed for virtual machines at different service classes. An analytic probabilistic model is then developed for non-steady VCDC states. In addition, a smart controller is developed for fine-grained resource provisioning and sharing among multiple applications. Furthermore, a novel dynamic hybrid metaheuristic algorithm is developed for the formulated profit maximization problem, based on simulated annealing and particle swarm optimization. The proposed algorithm can guarantee that differentiated service qualities can be provided with higher overall performance and lower energy cost. The advantage of the proposed approach is validated with trace-driven simulations.

Enforcing Differential Privacy for Shared Collaborative Filtering

Collaborative filtering is now successfully applied to recommender systems. The availability of extensive personal data is necessary for generating high quality recommendations. However, traditional collaborative filtering methods suffer from sparse and sometimes cold-start problems, particularly for newly deployed recommenders. Currently, several recommender systems exist in good working order, and data collected from these existing systems should be valuable for newly deployed recommenders. This paper introduces a privacy preserving shared collaborative filtering problem in order to leverage the data from other parties (contributors) to improve its own (beneficiaries) collaborative filtering performance, with the privacy protected under a differential privacy framework. It proposes a two-step methodology to solve this problem. First, item-based neighborhood information is selected as the shared data from the contributor with guaranteed differential privacy, and a practical enforcement mechanism for differential privacy is proposed. Second, two novel algorithms are developed to enable the beneficiary to leverage the shared data to support improved collaborative filtering. The extensive experimental results show that the proposed methodology can increase the recommendation accuracy of the beneficiary significantly while preserving data privacy for the contributors.

Time-Aware Multi-Application Task Scheduling With Guaranteed Delay Constraints in Green Data Center

A growing number of companies deploy their applications in green data centers (GDCs) and provide services to tasks of global users. Currently, a growing number of GDC providers aim to maximize their profit by deploying green energy facilities and decreasing brown energy consumption. However, the temporal variation in the revenue, price of grid, and green energy in tasks’ delay bounds makes it challenging for GDC providers to achieve profit maximization while strictly guaranteeing delay constraints of all admitted tasks. Unlike existing studies, a time-aware task scheduling (TATS) algorithm that investigates the temporal variation and schedules all admitted tasks to execute in GDC meeting their delay bounds is proposed. In addition, this paper provides the mathematical modeling of task refusal and service rates. In each iteration, TATS solves the formulated profit maximization problem by hybrid chaotic particle swarm optimization based on simulated annealing. Compared with several existing scheduling algorithms, TATS can increase profit and throughput without violating delay constraints of all admitted tasks. Note to Practitioners—This paper investigates the profit maximization problem for a green data center (GDC) while meeting delay constraints for all admitted tasks. Previous task scheduling algorithms do not jointly investigate temporal variation in revenue, green energy, and price of grid. Thus, they fail to meet the delay constraints of all admitted tasks. In this paper, a new approach that overcomes drawbacks of existing algorithms is proposed. It is obtained by using a hybrid metaheuristic algorithm that solves a constrained nonlinear optimization problem. Simulation results show that compared with several existing algorithms, it increases both throughput and profit. It can be readily incorporated into real-life industrial GDCs. The future work needs to investigate the repair/failure effect of GDCs on the proposed time-aware task scheduling.

A data-driven approach to predict Small-for-Gestational-Age infants

This work studies the problem of identifying risk factors of Small for Gestational Age (SGA) and building classifiers for SGA prediction. Recently, SGA infants have received more and more concerns as this illness brings many difficulties to them along with their whole life. Some experts have begun to study the risk factors of SGA onset by using traditional statistical ways. Others have used logistic regression (LR) to construct SGA prediction models. Meanwhile, machine learning have evolved and envisioned as a tool able to potentially identify babies with SGA. This work tests several feature selection methods. Based on the risk factors obtained through them, it trains support vector machine, random forest, and LR models and evaluates them via 10-fold cross validation in terms of precision and area under the curve of receiver operator characteristic curve. The results show that sparse LR of the wrapper algorithms owns the best feature selection effectiveness. In addition, this work compares data driven factors and knowledge driven factors and shows that the feature selection is necessary and effective. Among the trained classifiers, the LR model achieves the best performance on the data driven factors.

Pattern Recognition for Large-Scale and Incremental Time Series in Healthcare

In the big data era, large amounts of time series data have been collected from scientific experiments and business operations. With the fast growth of data volume, significant challenges in pattern recognition has emerged in diverse time series analysis systems. Since existing methods are inefficient and not applicable to large-scale and especially incremental time series, this paper proposes a novel and efficient pattern recognition method named MDLits for large-scale and incremental time series. It eliminates most redundant computation in the related works, and re-uses existing information by exploiting the correlation between existing data and newly-arrived ones. A Hadoop platform is implemented for clinical electrocardiography classification. The experiments on practical healthcare data show that our method outperforms the related arts in processing time, precision and recall. Moreover, it can scale to a large size and fit to incremental time series, which demonstrates the effectiveness and scalability of our method and system.

End-to-End Web Service Recommendations by Extending Collaborative Topic Regression

Mashuphasemergedasalightweightwaytocomposemultiplewebservicesandcreatevalue-added compositions.Facingthelargeamountofservices,effectiveservicerecommendationsareingreat need.Servicerecommendationsformashupqueriessuffersfromamashup-sidecold-startproblem, and traditional approaches usually overcome this by first applying topic models to mine topic proportionsofservicesandmashupqueries,andthenusingthemforsubsequentrecommendations. Thissolutionoverlooksthefactthatusagerecordcanprovideafeedbackfortextextraction.Besides, traditionalapproachesusuallytreatalltheusagerecordsequally,andoverlookthefactthattheservice usagepatternisevolving.Inthisarticle,theauthorsovercometheseissuesandproposeanend-toendservicerecommendationalgorithmbyextendingcollaborativetopicregression.Theresult is agenerativeprocesstomodelthewholeprocedureofserviceselection;thus,usagecanguidethe miningoftextcontent,andmeanwhile,theygivetime-awareconfidencelevelstodifferenthistorical usages.Experimentsonthereal-worldProgrammableWebdatasetshowthattheproposedalgorithm gainsanimprovementof6.3%intermsofmAP@50and10.6%intermsofRecall@50compared withthestate-of-the-artmethods. KEyWoRdS Mashup Development, Topic Modeling, Web Service Recommendations, Web Services

Cost-sensitive task routing and resource provisioning in geo-distributed clouds

Many different types of applications simultaneously execute in current data centers (DCs). To provide low cost and improved performance, each application is typically deployed in distributed DCs. Tasks of users around the world first go through Internet service providers (ISPs) which deliver data between distributed DCs and users. However, capacities and bandwidth cost of different ISPs vary. Besides, energy cost of multiple DCs located in different geographical places is different. With the growth of tasks, the DC providers energy and ISP bandwidth cost is huge and continues to increase. Therefore, due to the energy and bandwidth cost difference in different geographical places, it is highly difficult to minimize the DC providers total cost. Therefore, to tackle the problem, this work proposes a cost-sensitive task routing approach that can jointly specify the optimal selection of available ISPs for the arriving tasks, and the optimal number of switched-on servers in each DC. Finally, the simulation with tasks in Googles data center shows the proposed cost-sensitive task routing approach can effectively decrease the DC providers cost, and raise system throughput in comparison with some typical scheduling method.

Internet Communication Engine (ICE) Based Simulation Framework (ISF)

Most existing HLA based simulation frameworks do not support simulation on WAN. To tackle this problem, we propose an Internet Communication Engine (ICE) based framework (ISF) that is capable of running on WAN, generating simulation codes automatically, and managing simulation procedures efficiently. ICE as well as its several powerful services is adopted to build discrete event driven applications and provide network communications. A use case of antimissile system is implemented with this framework, which is validated to have useful functions and be stable with running simulations on both LAN and WAN.