• Journal of Digital Contents Society
• /
• v.2 no.2
• /
• pp.167-178
• /
• 2001

For high performance Grid environments, an effective GIS(Grid Information System) should be adopted. In the design of GIS architecture, its grid resource information model provides a key basis. This paper presents our study on the exploration of a high performance grid resource information model. According the exploration, we identified the followings. The resource information model should clearly address the issues of relation descriptions as well as resource descriptions, issues related to scheduling support, the issue of decoupling the expression model of resource information from data repository models, and finally the issue of decoupling user-level resource descriptions from system-level resource descriptions. Based on the proposed conceptual organization of resource information models, analysis result for the existing resource information models are presented.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.3
• /
• pp.67-77
• /
• 2024

Increasing energy efficiency in factories is an activity aimed at optimizing resource allocation in manufacturing processes to establish production plans. However, this strategy may not apply effectively when night shifts are unavoidable. Additionally, continuous fluctuations in production requirements pose challenges for its implementation in the factory. Recently, with the rapid proliferation of electric vehicles (EVs), technology utilizing electric vehicle batteries as energy storage systems has gained attention. Technology using these batteries can be an alternative for factory energy management. In this paper, a factory energy management method using EV batteries is proposed. The proposed method is analyzed using PSCAD/EMTDC software, considering the state of charge of EV batteries and Time-of-Use (TOU) rates. The proposed method was compared with production scheduling established considering predicted power usage and TOU rates. As a result, production scheduling saved 4,152 KRW per day, while the proposed method saved 7,286 KRW in electricity costs. Through this paper, the possibility of utilizing EV batteries for factory energy management has been demonstrated.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.11a
• /
• pp.171-172
• /
• 2009

가상화 기술을 기반으로 하는 클라우드 컴퓨팅은 시스템 내에 존재하는 자원을 분할하여 이용하거나 독립적으로 존재하는 자원을 통합하는 측면에서는 매우 편리하고 용이하다. 또한 컴퓨팅 자원의 활용률을 극대화하는 이점도 존재한다. 하지만 가상화 기술 자체만으로는 사용자의 요청에 동적으로 가상화된 자원, 즉 가상 머신(Virtual Machine)을 생성, 복제, 할당 및 삭제 등의 관리 작업에 어려움이 존재한다. 따라서 본 논문에서는 Xen 과 Globus Toolkit 을 기반으로 구축된 환경에서 가상 머신의 관리를 편리하게 처리할 수 있는 가상 머신 매니저를 그리드 서비스로 설계 및 구현하였다. 또한 사용자의 요청을 효율적으로 처리하기 위해 작업과 자원에 대한 실시간 스케줄링 방법을 제안한다.

• Journal of Korea Game Society
• /
• v.16 no.4
• /
• pp.57-68
• /
• 2016

As the size of the data used in the game application increase gradually, the physical resources of game server grow. Accordingly, it is necessary to/it is required to introduce I/O virtualization in game server to improve I/O performance of the server. But it is difficult to ensure high responsiveness in game server where I/O delay change from time to time. To maximize the benefit of I/O virtualization, guaranteeing I/O response time is very important and it is necessary to have I/O scheduling techniques to manage the I/O latency in the order of priority of virtual machines. In this paper, we propose an efficient delay-constrained scheduling algorithm in a virtualization environment to ensure maximum I/O latency. In addition, a reduced amount of loss of the packet was found to increase the fairness of scheduling in the experiments with the proposed scheme.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.770-771
• /
• 2015

In this paper, we proposed and implemented the automation of cable drum scheduling which is required a considerable construction cost for electrical equipment in power plant. Implemented cable drum scheduling can be reduced the calculation time of cable drum scheduling, and it is possible to minimize the consuming time of cable drum scheduling than conventional methods. Some experiments are conducted to verify the proposed method, and as a result, the automation of cable drum scheduling is well performed, enabling cables can be optimized and scheduled to fit the drum capacity with productivity enhancement of electrical installations.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.6A
• /
• pp.558-564
• /
• 2007

3G Long Term Evolution, which aims for various mobile multimedia service provision by enhanced wireless interface, proposes VoIP-based voice service through a Packet Switching (PS) domain. As delay and loss-sensitive VoIP traffic flows through the PS domain, more challenging technical difficulties are expected than in Circuit Switching (CS) domain based VoIP services. Moreover, since 3G LTE, which adopts the OFDM as its physical layer, introduces Physical Resource Block (PRB) as a unit for transmission resources, new types of resource management schemes are needed. This paper proposes a PRB scheduling algorithm of MAC layer for VoIP service in 3G LTE and shows the simulation results. The proposed algorithm has two key parts; dynamic activation of VoIP priority mode to satisfy VoIP QoS requirements and adaptive adjustment of the priority mode duration in order to minimize the degradation of resource utilization.

• The KIPS Transactions:PartC
• /
• v.11C no.4
• /
• pp.555-562
• /
• 2004

We consider differentiated bandwidth allocation for a piconet in short-range wireless personal network systems : Bluetooth. Since bandwidth requirements nay vary among applications/services, and/or it may change over time, it is important to decide how to allocate limited resources to various service classes to meet their service requirements. We propose a simple and efficient bandwidth allocation mechanism which meets bandwidth requirements of various service types while saving power consumption by a Power saying mode, i.e., sniff node. We compare our proposed mechanism with a conventional (weighted) round-robin polling scheme and show that it achieves significant improvement of hroughput, delay, and power consumption.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.2
• /
• pp.71-76
• /
• 2022

Round-robin is widely used for the scheduling of large-scale parallel workloads in the computing units of GPGPU. Round-robin is easy to implement by sequentially allocating tasks to each computing unit, but the load balance between computing units is not well achieved in multi-workload environments like cloud. In this paper, we propose a new thread block scheduling policy to resolve this situation. The proposed policy manages thread blocks generated by various GPGPU workloads with multiple queues based on their computation loads and tries to maximize the resource utilization of each computing unit by selecting a thread block from the queue that can maximally utilize the remaining resources, thereby inducing load balance between computing units. Through simulation experiments under various load environments, we show that the proposed policy improves the GPGPU performance by 24.8% on average compared to Round-robin.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.1A
• /
• pp.13-23
• /
• 2007

In this paper, a packet scheduling scheme that supports real-time traffic having multi-level delay constraints in OFDMA systems is proposed. The proposed scheme pursues to satisfy the delay constraint first, and then manage the residual radio resource in order to enhance the overall throughput. A parameters named tolerable delay time (TDT) is newly defined to deal with the differentiated behaviors of packet scheduling according to the delay constraint level. Assuming that the packets violating the delay constraint are discarded, the proposed scheme is evaluated in terms of the packet loss probability, throughput, channel utilization. It is then compared with existing schemes for real-time traffic support such as the Exponential Scheduling (EXP) scheme, the Modified Largest Weighted Delay First (M-LWDF) scheme, and the Round robin scheme. The numerical results show that the proposed scheduling scheme performs much better than the aforementioned scheduling schemes in terms of the packet loss probability, while slightly better in terms of throughput and channel utilization.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.04a
• /
• pp.707-710
• /
• 2002

실시간 태스크는 제한된 시간 내에 작업이 수행되어야 하는 특성을 가지고 있으며, 이러한 특성을 만족시켜주기 위해서 필요한 자원이 확보되어 있어야 한다. 본 논문에서는 제한 시간이 완화된 연성 실시간 태스크를 수행시킴에 있어서 기존에 제시된 메모리 관리 기법의 문제점을 분석하고, 연성 실시간 태스크가 원활히 수행될 수 있도록 하는 방법을 제시하고 있다 본 논문에서 제시하는 방법은 프로세스 스케줄러와 메모리 관리 스케줄러가 상호 협조적으로 동작하는 것에 초점을 맞추고 있다 즉, 프로세스 스케줄링 정보와 물리 메모리 사용량을 기초로 연성 실시간 태스크가 필요로 하는 메모리 자원이 page-out 되거나 swap-out 되는 비율을 조정하고 있다 제시된 방법을 실험을 통하여 검증을 하였으며, 실험 결과에서는 물리 메모리의 부족으로 인한 과부하 상황에서 연성 실시간 태스크의 제한 시간 실패율을 감소되고 전체적인 성능이 향상됨을 보이고 있다.