• International journal of advanced smart convergence
• /
• v.9 no.3
• /
• pp.253-259
• /
• 2020

Most of today's large-scale cloud systems and enterprise data centers are distributing resources to improve scalability and resource utilization. NVMe-over-Fabric protocol allows submitting NVMe commands to a remote NVMe SSD through RDMA (Remote Direct Memory Access) network. It is attracting attention recently because it is possible to construct a disaggregation storage system with low latency through the protocol. However, the current I/O stack of NVMe-over-Fabric has an inefficient structure for maintaining compatibility with the traditional I/O stack. Therefore, in this paper, we propose a new mechanism to reduce I/O latency and CPU overhead by modifying I/O path of NVMe-over-Fabric to pass through legacy block layer. According to the performance evaluation results, the proposed mechanism is able to reduce the I/O latency and CPU overhead by up to 22% and 24% compared to the existing NVMe-over-Fabrics protocol, respectively.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1706-1711
• /
• 2008

This paper proposes a new Web based-control concept & design method and implementation of LonWorks network system for a large scale renewable energy energy control and monitoring system in building. The Experimental LonWorks network system using LonWorks/Ethernet(Web) server is designed and fabricated. This article addresses issues in architecture of LonWorks/Ethernet sever, embedded processors architecture for converting LonTalks protocol to Modbus protocol and software, and Internet technologies. It is also verified that the multi-induction motor control and monitoring system using LonWorks/Ethernet server have available, interoperable, reliable performance characteristics from the experimental results, Especially, The seamless integration of TCP/IP networks with control networks allows access to any control point from anywhere. Thus, the results provide a available technical data for remote distributed control system of industrial and buildings field.

• Journal of the Korea Computer Industry Society
• /
• v.2 no.9
• /
• pp.1225-1232
• /
• 2001

In this paper, we investigate the scheduling algorithm of router system for Internet services on large scale ATM networks based on the quality-of-service(QoS) level of the input source traffics. We suggest an approprite scheduling algorithm in order to satisfy their QoS requirements. For this purpose, we first study the service requirements of the multiplexer in Internet. Second, we suggest functional architecture of the multiplexer for real time services and the scheduling algorithm to satisfy various QoS requirements. Finally, the performance measures of interest, namely steady-state average delay time and fairness of network resources, are discussed by simulation results.

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06c
• /
• pp.377-378
• /
• 2011

본 논문은 뇌기능 연구에 크게 기여하는 기능적 자기공명영상을 효과적으로 분석하기 위한 유효 연결성(Effective Connectivity, EC)을 이용한 대규모 네트워크(Large-Scale Network, LSN) 분석(LSN-EC)을 제안한다. 유효 연결성은 뇌영역간의 시공간적 인과관계를 표현한 연결성이며, 뇌의 기능적 연결성 및 구조탐색 사용된다. LSN-EC는 뇌영역간의 EC를 표현하고 그룹간의 차이분석을 통하여 뇌질환 분석 및 진단 연구로의 응용이 가능하다. 실험결과에서 알츠하이머병과 관련이 높다고 알려진 후대상피질(Posterior Cingulate Cortex)과 해마(Hippocampus)가 포함된 변연엽(Limbic Lobe), 기저핵 및 시상(Basal Ganglion and Thalamus) 주변 영역에서 감소된 EC를 확인하였다.

• Journal of Communications and Networks
• /
• v.17 no.5
• /
• pp.473-481
• /
• 2015

In this study, we propose a clustering algorithm to enhance the performance of wireless sensor and actuator networks (WSANs). In each cluster, a multi-level hierarchical structure can be applied to reduce energy consumption. In addition to the cluster head, some nodes can be selected as intermediate nodes (INs). Each IN manages a subcluster that includes its neighbors. INs aggregate data from members in its subcluster, then send them to the cluster head. The selection of intermediate nodes aiming to optimize energy consumption can be considered high computational complexity mixed-integer linear programming. Therefore, a heuristic lowest energy path searching algorithm is proposed to reduce computational time. Moreover, a channel assignment scheme for subclusters is proposed to minimize interference between neighboring subclusters, thereby increasing aggregated throughput. Simulation results confirm that the proposed scheme can prolong network lifetime in WSANs.

• 전기의세계
• /
• v.30 no.5
• /
• pp.297-305
• /
• 1981

The problem of finding shortest paths between every pair of points in a network is solved employing and optimal network decomposition in which the network is decomposed into a number of subnetworks minimizing the number of cut-set between them while each subnetwork is constrained by a size limit. Shortest path computations are performed on individual subnetworks, and the solutions are recomposed to obtain the solution of the original network. The method when applied to large scale networks significantly reduces core requirement and computation time. This is demonstrated by developing a computer program based on the method and applying it to 30-vertex, 160-vertex, and 273-vertex networks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.551-569
• /
• 2024

With the rapid development of electric vehicles (EVs) industry, EV charging service becomes more and more important. Especially, in the case of suddenly drop of air temperature or open holidays that large-scale EVs seeking for charging devices (CDs) in a short time. In such scenario, inefficient EV charging scheduling algorithm might lead to a bad service quality, for example, long queueing times for EVs and unreasonable idling time for charging devices. To deal with this issue, this paper propose a Deep-Q-Network (DQN) based two-stage scheduling method for the large-scale EVs charging service. Fine-grained states with two delicate neural networks are proposed to optimize the sequencing of EVs and charging station (CS) arrangement. Two efficient algorithms are presented to obtain the optimal EVs charging scheduling scheme for large-scale EVs charging demand. Three case studies show the superiority of our proposal, in terms of a high service quality (minimized average queuing time of EVs and maximized charging performance at both EV and CS sides) and achieve greater scheduling efficiency. The code and data are available at THE CODE AND DATA.

• Journal of KIISE
• /
• v.41 no.9
• /
• pp.686-698
• /
• 2014

Recently, large-scale streaming sensor data have emerged due to explosive supply of smart phones, diffusion of IoT and Cloud computing technology, and generalization of IoT devices. Also, researches on combination of semantic web technology are being actively pushed forward by increasing of requirements for creating new value of data through data sharing and mash-up in large-scale environments. However, we are faced with big issues due to large-scale and streaming data in the inference field for creating a new knowledge. For this reason, we propose the RDFS rule based parallel reasoning scheme to service by processing large-scale streaming sensor data with the semantic web technology. In the proposed scheme, we run in parallel each job of Rete network algorithm, the existing rule inference algorithm and sharing data using the HBase, a hadoop database, as a public storage. To achieve this, we implement our system and evaluate performance through the AWS data of the weather center as large-scale streaming sensor data.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2008.10a
• /
• pp.307-308
• /
• 2008

• Proceedings of the Korean Nuclear Society Conference
• /
• 1998.10a
• /
• pp.87-87
• /
• 1998