• Journal of Information Processing Systems
• /
• v.18 no.6
• /
• pp.803-812
• /
• 2022

With the increasing number of mobile device users worldwide, utilizing mobile edge computing (MEC) devices close to users for content caching can reduce transmission latency than receiving content from a server or cloud. However, because MEC has limited storage capacity, it is necessary to determine the content types and sizes to be cached. In this study, we investigate a caching strategy that increases the hit ratio from small base stations (SBSs) for mobile users in a heterogeneous network consisting of one macro base station (MBS) and multiple SBSs. If there are several SBSs that users can access, the hit ratio can be improved by reducing duplicate content and increasing the diversity of content in SBSs. We propose a Deep Q-Network (DQN)-based caching strategy that considers time-varying content popularity and content redundancy in multiple SBSs. Content is stored in the SBS in a divided form using maximum distance separable (MDS) codes to enhance the diversity of the content. Experiments in various environments show that the proposed caching strategy outperforms the other methods in terms of hit ratio.

• Journal of the Korea Society for Simulation
• /
• v.18 no.2
• /
• pp.29-38
• /
• 2009

In the computing environment with heterogeneous resources, a job scheduling model is necessary for effective resource utilization and high-speed data processing. And, the job scheduling model has to cope with a dynamic change in the condition of resources. There have been lots of researches on resource estimation methods and heuristic algorithms about how to distribute and allocate jobs to heterogeneous resources. But, existing researches have a weakness for system compatibility and scalability because they do not support the standard language. Also, they are impossible to process jobs effectively and deal with a variety of computing situations in which the condition of resources is dynamically changed in real-time. In order to solve the problems of existing researches, this paper proposes a semantic computing-based dynamic job scheduling model that defines various knowledge-based rules for job scheduling methods adaptable to changes in resource condition and allocate a job to the best suited resource through inference. This paper also constructs a resource ontology to manage information about heterogeneous resources without difficulty as using the OWL, the standard ontology language established by W3C. Experimental results shows that the proposed scheduling model outperforms existing scheduling models, in terms of throughput, job loss, and turn around time.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.637-640
• /
• 2017

최근 클라우드 기술이 확산되면서 여러 기업이 자신만의 클라우드를 다양하게 제공하고 있으며 세계 각지에 흩어져 있는 다양한 클라우드 공급자들과 여러 프라이빗 클라우드를 연결하여 하나의 서비스를 제공하는 클라우드 서비스 브로커리지(Cloud Services Brokerage) 기술이 등장하게 되었다. 하지만 현재의 CSB 기술은 가상 자원 관리, 가상 머신 라이프 사이클 관리에 초점이 맞추어져 있으며 CSB에서 제공되는 모니터링 요소는 사용자의 복잡한 요구를 만족하기에는 부족한 실정이다. 또한 추가 모니터링 도구를 사용하는 경우도 CSB의 데이터와의 일관성을 만족하지 않는 문제가 있다. 따라서 본 논문에서는 이종 CSB와 연동이 가능한 실시간 모니터링 시스템을 제안한다. 본 논문에서 제안하는 모니터링 기술은 Scalr 및 CompatibleOne을 사용하여 주기적으로 메타데이터 동기화를 진행하며 데이터 일관성을 충족시키고 다양한 모니터링 데이터를 실시간으로 수집하여 사용자에게 제공한다.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.1
• /
• pp.20-30
• /
• 2013

An intelligent grid computing architecture is proposed and developed for transient stability constrained total transfer capability evaluation of future smart grid. In the proposed intelligent grid computing architecture, a model of generalized compute nodes with 'able person should do more work' feature is presented and implemented to make full use of each node. A timeout handling strategy called conditional resource preemption is designed to improve the whole system computing performance further. The architecture can intelligently and effectively integrate heterogeneous distributed computing resources around Intranet/Internet and implement the dynamic load balancing. Furthermore, the robustness of the architecture is analyzed and developed as well. The case studies have been carried out on the IEEE New England 39-bus system and a real-sized Chinese power system, and results demonstrate the practicability and effectiveness of the intelligent grid computing architecture.

• The Journal of the Korea Contents Association
• /
• v.5 no.1
• /
• pp.189-199
• /
• 2005

The grid computing is a technique to solve big problems such as a field of scientific technique by sharing the computing power and a big storage space of the numerous computers on the distributed network. The environment of the grid computing is composed with the WAN which has a different performance and a heterogeneous network condition. Therefore, it is more important to reflect heterogeneous performance elements to calculation work. In this paper, we propose an efficient method that decides work process number of each node by considering a network state information. The network state information considers the latency, the bandwidth and latency-bandwidth mixture information. First, using information which was measured, we compute the performance ratio and decide work process number of each node. Finally, RSL file was created automatically based on work process number which was decided, and then accomplishes a work. The network performance information is collected by the NWS. According to experimental results, the method which was considered of network performance information is improved respectively 23%, 31%, and 57%, compared to the methods of existing in a viewpoint of work amount, work process number, and node number.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.731-763
• /
• 2017

In Underwater Linear Sensor Networks (UW-LSN) routing process, nodes without proper address make it difficult to determine relative sensor details specially the position of the node. In addition, it effects to determine the exact leakage position with minimized delay for long range underwater pipeline monitoring. Several studies have been made to overcome the mentioned issues. However, little attention has been given to minimize communication delay using dynamic addressing schemes. This paper presents the novel solution called Hop-by-Hop Dynamic Addressing based Routing Protocol for Pipeline Monitoring (H2-DARP-PM) to deal with nodes addressing and communication delay. H2-DARP-PM assigns a dynamic hop address to every participating node in an efficient manner. Dynamic addressing mechanism employed by H2-DARP-PM differentiates the heterogeneous types of sensor nodes thereby helping to control the traffic flows between the nodes. The proposed dynamic addressing mechanism provides support in the selection of an appropriate next hop neighbour. Simulation results and analytical model illustrate that H2-DARP-PM addressing support distribution of topology into different ranges of heterogeneous sensors and sinks to mitigate the higher delay issue. One of the distinguishing characteristics of H2-DARP-PM has the capability to operate with a fewer number of sensor nodes deployed for long-range underwater pipeline monitoring.

• Journal of Convergence for Information Technology
• /
• v.9 no.9
• /
• pp.33-37
• /
• 2019

According to the enforcement of the Cloud Computing Development and User Protection Act, a new paradigm called as Cloud Computing is coming to the fore among public and private domestic enterprises. Therefore, domestic companies should develop SaaS products specialized for the domestic market and try to preempt the IaaS market ahead of global companies. Now we are facing a necessity to combine all the cloud systems in Korean government to operate seamlessly in a harmony. In this paper, we will look at the prospect of future development of related industry through cloud computing concept and G-cloud's cloud management system (G-CMS). G-CMS can be seen as the first system to comprehensively manage heterogeneous Unix virtualization systems. G-CMS can also save costs by managing heterogeneous virtualization systems in data centers.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.2
• /
• pp.237-246
• /
• 1995

As the needs for accessing distributed informations and computing resources are increasing the need for network interconnection is growing. There are many of the network management packages including performance management tools; but most of them are not appropriate to be used in heterogeneous interconnected networks. To monitor and to control efficiently the performance of heterogeneous network, first, we have to define all performance parameters in general meaning. We need models and criteria for supporting performance analysis activities of the management system. In this study, we have designed a centralized performance management system based on the OSI management, which can be used in heterogeneous networks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.10
• /
• pp.4642-4660
• /
• 2016

One major concern of applying Multipath TCP (MPTCP) to data delivery in heterogeneous wireless networks is that the utilization of asymmetric paths with diverse networking-related parameters may cause severe packet reordering and receive buffer blocking (RB²LOC). Although many efforts are devoting to addressing MPTCP's packet reordering problems, their sender-controlled solutions do not consider balancing overhead between an MPTCP sender and receiver, and their fully MPTCP mode cannot make MPTCP achieve a desired performance. This paper proposes a novel receiver-centric buffer blocking-aware data scheduling strategy for MPTCP (dubbed MPTCP-rec) necessitating the following aims: (1) alleviating MPTCP's packet reordering and RB²LOC problems, (2) improving the MPTCP performance, and (3) balancing load between the MPTCP sender and receiver. Simulation results show that the proposed MPTCP-rec solution outperforms the existing MPTCP solutions in terms of data delivery performance in heterogeneous wireless networks.

• Coupled systems mechanics
• /
• v.11 no.2
• /
• pp.167-198
• /
• 2022

In this paper we deal with classical instability problems of heterogeneous Euler beam under conservative loading. It is chosen as the model problem to systematically present several possible solution methods from simplest deterministic to more complex stochastic approach, both of which that can handle more complex engineering problems. We first present classical analytic solution along with rigorous definition of the classical Euler buckling problem starting from homogeneous beam with either simplified linearized theory or the most general geometrically exact beam theory. We then present the numerical solution to this problem by using reduced model constructed by discrete approximation based upon the weak form of the instability problem featuring von Karman (virtual) strain combined with the finite element method. We explain how such numerical approach can easily be adapted to solving instability problems much more complex than classical Euler's beam and in particular for heterogeneous beam, where analytic solution is not readily available. We finally present the stochastic approach making use of the Duffing oscillator, as the corresponding reduced model for heterogeneous Euler's beam within the dynamics framework. We show that such an approach allows computing probability density function quantifying all possible solutions to this instability problem. We conclude that increased computational cost of the stochastic framework is more than compensated by its ability to take into account beam material heterogeneities described in terms of fast oscillating stochastic process, which is typical of time evolution of internal variables describing plasticity and damage.