• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06b
• /
• pp.464-469
• /
• 2007

현재 이동용 장치(Mobile Device)들에서 전력 소모는 사용자들의 요구에 따라 성능 다음으로 중요한 비중을 차지하고 있다. 특히 배터리 셀의 기술 증가에 비해 프로세서들의 성능 및 요구하는 소비전력이 크게 증가함에 따라 프로세서의 전력 소모를 최소화 하는 연구들이 많이 진행되고 있다. 특히 프로세서의 전력 소모가 많은 비중을 차지함에 따라 프로세서의 전력 소모를 낮추기 위한 방법으로 많은 프로세서들은 DVS(Dynamic Voltage Scaling)와 DFS(Dynamic Frequency Scaling)를 지원한다. 실제 프로세서의 전력 소모는 공급전압에 의 제곱에 비례하고 동작 클럭(Clock) 주파수에 비례한다. 그러나 공급전압은 다시 동작 클럭 주파수에 비례함으로써 DVS와 DFS를 지원하는 대부분의 프로세서는 동작 클럭 주파수를 낮춤으로서 많은 전력 소모를 줄일 수 있게 된다. 그러나 동작 클럭 주파수를 낮추게 되면 태스크들의 실행 시간이 길어지게 되어 실시간 시스템에서 실시간성을 보장하지 못하게 된다. 본 논문에서는 상호간에 공유자원을 갖는 태스크들의 실시간성을 보장하며 동작 클럭 주파수를 낮추는 알고리즘을 제안한다.

• The KIPS Transactions:PartA
• /
• v.14A no.6
• /
• pp.357-362
• /
• 2007

In distributed systems like cluster systems, in order to get more performance improvement, the initial task placement system precisely estimates and correctly assigns the resource requirement by the process. The resource-based initial job placement scheme needs the prediction of resource usage of a task in order to fit it to the most suitable hosts. However, the wrong prediction of resource usage causes serious performance degradation in dynamic load balancing systems. Therefore, in this paper, to resolve the problem due to the wrong prediction, we propose a new load metric. By the new load metric, the resource-based initial job placement scheme can work without priori knowledge about the type of process. Simulation results show that the dynamic load balancing system using the proposed approach achieves shorter execution times than the conventional approaches.

• Journal of KIISE:Information Networking
• /
• v.33 no.2
• /
• pp.183-192
• /
• 2006

This paper analyzes the price-selection problem under priority-based scheduling for QoS (Quality of Service) network services, i.e., how to determine the price associated with each service level. In particular, we focus on the problems with the pricing mechanism based on equilibrium analysis. We claim that the assumptions needed to produce equilibrium nay not hold in some important environments. Specifically, (a) the individual user's impact on the system is not infinitesimal and (b) users do not always have up-to-date global system-status knowledge crucial for optimal user decisions required for equilibrium. These may make the equilibrium models inaccurate in realistic environments. We examine the accuracy of some existing equilibrium methods by using a dynamic model that we have developed for system behavior analysis. The analysis indicates that equilibrium methods fail to model accurately the system behavior in some realistic environments.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.3A
• /
• pp.259-266
• /
• 2011

Femtocells are expected as the surest way to increase the system capacity with higher-quality links and more spatial reuse in future networks. In spite of their great potential, the system capacity is highly susceptible to network density because a large portion of users are exposed to inter-cell interference (ICI). In this work, we proposed a dynamic interference avoidance scheme in densely deployed cell environments. Our proposed DDIA (Distributed Dynamic ICI Avoidance) scheme not only works in a fully distributed manner, but also controls interference link connectivity of users with high agility so that it is suited for self-organizing networks (SONs). We introduced the concept of ICI-link and two-tier scheduling in designing the DDIA scheme. To avoid ICI without any central entity, our scheme tries to harmonize all base stations (BSs) with users adaptively. Through extensive simulations, it was shown that our proposed scheme improves the throughput of users by more than twice on average compared to the frequency reuse factor 1 scheme, who are exposed to ICI while maintaining or even improving overall network performance. Our scheme operates well regardless of network density and topology.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.34 no.5
• /
• pp.193-200
• /
• 1985

Increased attention has been given in recent years to the use of pumped hydrostorage plant to meet the requirements for peak generation and bottom pumping. Once a pumped hydrostorage plant is installed, its economic operation as an integral component of a steam generating system requires the selection of a pumping and generating schedule which will result in the most effective use of the hydro generating capacity. The general object of coordination of pumped hydrostorage plants with electric power system is the minimization of the overall procduction cost and the maximization of generation reserves. This paper presents a method for the optimal scheduling of pumped hydrostorage plant and a computer program which determines weekly operating schedules for a pumped hydrostorage plant by dynamic programming method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.1
• /
• pp.117-135
• /
• 2016

Performance enhancement is one of the most important issues in high performance distributed computing systems. In such computing systems, online users submit their jobs anytime and anywhere to a set of dynamic resources. Jobs arrival and processes execution times are stochastic. The performance of a distributed computing system can be improved by using an effective load balancing strategy to redistribute the user tasks among computing resources for efficient utilization. This paper presents a multi-class load balancing strategy that balances different classes of user tasks on multiple heterogeneous computing nodes to minimize the per-class mean response time. For a wide range of system parameters, the performance of the proposed multi-class load balancing strategy is compared with that of the random distribution load balancing, and uniform distribution load balancing strategies using simulation. The results show that, the proposed strategy outperforms the other two studied strategies in terms of average task response time, and average computing nodes utilization.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.1B
• /
• pp.17-21
• /
• 2007

In this paper, we propose new on-line processor management algorithms that manage heterogeneous multimedia services while maximizing energy efficiency. These online management mechanisms are combined in an integrated scheme for higher system performance and energy efficiency. The most important feature of our proposed scheme is its adaptability, flexibility and responsiveness to current network conditions. Simulation results clearly indicate the superior performance of our proposed scheme to strike the appropriate performance balance between contradictory requirements.

• Proceedings of the KIEE Conference
• /
• 1999.07a
• /
• pp.101-104
• /
• 1999

This paper presents the design of state feedback controller with state observer(dynamic filter) for Electro-Magnetic Suspension System. Also, the gain scheduling technique using state observer is analyzed in the time domain and frequency domain.

• Journal of KIISE:Computer Systems and Theory
• /
• v.28 no.11
• /
• pp.549-560
• /
• 2001

Cluster system provides attractive scalability in terms of compution power and memory size. With the advances in high speed computer network technology, cluster systems are becoming increasingly competitive compared to expensive MPPs (massively parallel processors). Load balancing is very important issue since an inappropriate scheduling of tasks cannot exploit the true potential of the system and can offset the gain from parallelization. In parallel processing program, it is difficult to predict the load of each task before running the program. Furthermore, tasks are interdependent each other in many ways. The dynamic load balancing algorithm, which evaluates each processor's load in runtime, partitions each task into the appropriate granularity and assigns them to processors in proportion to their performance in cluster systems. However, if the communication cost between processing nodes is expensive, it is not efficient for all nodes to attend load balancing process. In this paper, we restrict a processor that attend load balancing by the communication cost and the deviation of its load from the average. We simulate various models of the cluster system with parameters such as communication cost, node number, and range of workload value to compare existing load balancing methods with the proposed dynamic algorithms.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.45 no.9
• /
• pp.39-48
• /
• 2008

Recently, many multimedia services have emerged in Internet such as real-time and non- real time services. However, in this Internet environment, we have some limitations to satisfy each service feature. To guarantee the service features in Measurement-Based Admission Control(MBAC) based system on the flow-aware network, there is the method applying Dynamic Priority Scheduling(DPS) algorithm that gives a higher priority to an earlier incoming flow in all of the link bandwidth. This paper classifies all flows under several groups according to flow characteristics on per-flow MBAC algorithm based system. In each flow group, DPS algorithm is applied. This paper proposes two methods that are a DPS based bandwidth borrowing method and a bandwidth dynamic allocation method. The former is that if low priority class has available bandwidths, the flow of high priority class borrows the bandwidth of low priority class when high priority flow has insufficient bandwidth to connect a flow call. The later is that the each group has a minimum bandwidth and is allocated the bandwidth dynamically according to the excess rate for available bandwidth. We compare and analyze the characteristics of the two proposed methods through the simulation experiments. As the results of the experiment, the proposed methods are more effective than existing DPS based method on the packet loss and delay characteristics. Consequently the proposed two methods are very useful in various multimedia network environments.