A5100344831- Cloud Computing and Resource Management
- IoT and Edge/Fog Computing
- Advanced Decision-Making Techniques
- Technology and Security Systems
- Caching and Content Delivery
- Cloud Data Security Solutions
- Web and Library Services
- Digital and Cyber Forensics
- Distributed and Parallel Computing Systems
- Privacy-Preserving Technologies in Data
- Cybersecurity and Information Systems
- Cybercrime and Law Enforcement Studies
- Graph Theory and Algorithms
- Educational Technology and Assessment
- QR Code Applications and Technologies
- Big Data Technologies and Applications
- Age of Information Optimization
- Acoustic Wave Phenomena Research
- UAV Applications and Optimization
- Internet of Things and AI
- Evaluation Methods in Various Fields
- Distributed systems and fault tolerance
- Crime Patterns and Interventions
- Data Quality and Management
- Smart Grid and Power Systems
Hunan University of Science and Technology
2019-2023
NetEase (China)
2015-2021
University of Science and Technology of China
2019
Zhejiang Lab
2019
Tsinghua University
2014
Wuhan Textile University
2012
IBM (United States)
2010
University of Arizona
2006
With the popularity of container-based microservices and cloud-native architectures, Kubernetes has established itself as de facto standard for container orchestration. is known its advantage in easy deployment operations applications; however, it suffers from low resource utilization, incurring high server provisioning operational costs, due to following reasons. First, common practice latency-sensitive services be over-provisioned peak load: might consider such peak-load constant, even...
Summary Resource usage prediction is increasingly important in cloud computing environments, and CPU especially helpful for improving the efficiency of resource provisioning reducing energy consumption datacenters. However, accurate remains a challenge few works have been done on predicting physical machines In this article, we present deep belief network (DBN) particle swarm optimization (PSO) based algorithm, which named DP‐CUPA aimed to provide more results. The consists three main steps....
Summary To meet the ever‐increasing requirements of applications, cloud service providers have further built and managed multiple data centers in regions across geographies with many physical machines (PMs). However, most existing resource allocation algorithms are developed for a single center, which normally cannot efficiently handle load burst occasions where center may not be enough to satisfy demands applications. Therefore, it is necessary consider how manage while meeting application...
With the rapid development of Internet Things and ever-increasing demands advanced services applications, edge computing is proposed to move storage resources near data source, which improves response time saves bandwidth. However, due limited available massive privacy-sensitive user in nodes, there are huge challenges security privacy protection environment. Hence, we propose an efficient time-domain multi-authority outsourcing attribute-based encryption (ABE) scheme (TMO) with a dynamic...
The goal of the COPLINK project is to develop information and knowledge management systems technologies methodology appropriate for capturing, accessing, analyzing, visualizing, sharing law enforcement related information. Continuing our pursuit this goal, we have developed a framework creating cross-jurisdictional Criminal Activity Networks social network analysis, constructed taxonomy criminal identity resolution problems detecting intentional deception unintentional errors.
A cloud-based, highly consumable queuing service must provide extreme scalability, flexible models of consistency, and high availability in the presence network partitions. CAP theorem states that at most two three properties - Consistency, Availability, Partition Tolerance can be achieved same time for any shared data system. This paper presents design implementation a scalable cloud-based service, named Silver Dove Queuing Service (SDQS). SDQS offers kinds consistency levels to...
Aiming at low portability and undue dependence on the SSD database of most authentication system, a fingerprint identification system based two-dimensional bar code DSP was proposed. The central processor TMS320VC5416 used, information obtained from PDF417 barcode by reading device fuzzy matched with that online real-time collection. quick verification card holder’s identity realized.
In recent ten years, the problems related to voltage stability arise frequently worldwide. The intelligent algorithm, represented by fuzzy logic, neural network and genetic has many applications in analysis of problems. This paper will mainly introduce such application.
The development of cloud computing has promoted the predictive technology. A number prediction methods have been proposed, but these require a lot manual operations to find appropriate parameters. This caused great inconvenience. Therefore, this paper designs and implements tool that allows users select models independently test optimal combination RPT is an openly improvable toolbox permits directly establish elastic offline resource forecast system in which resources are furnished by user....
With the continuous development of spaceflight industry in recent years, number on-orbit satellites is rapidly increasing, scale Satellite Control Center (SCC) becoming larger and larger, cloud technology has been applied completely. Although plays an important role on improving system flexibility, reliability recoverability SCC, there are still some insufficiency to be solved at present, such as virtual host configuration yet complex, complexity difficulty maintenance increasing more kinds...
With the wide application of motor control systems, performance has become more stable. However, in face increasingly complex electromagnetic environment, compatibility problems systems and severe. This article takes system composed stepping as research object. experimental demonstration, sensitivity mechanism stepper under typical interference signals is studied, threshold law model are proposed. sensitive related to length time power signal.