• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.11a
• /
• pp.678-681
• /
• 2007

분산형 데스크탑 그리드 시스템에서 안정적인 연산 수행을 위한 노드 구성 기법과 동적인 환경에 적응적인 스케줄링 기법은 필수 요소이다. 그러나 기존 연구에서는 자원 제공자의 휘발성에 적응적으로 대처하지 못하는 연산 수행 모델을 사용하였기 때문에 데드라인 내에 전체 작업을 완료해야하는 응용에 대처하지 못하는 문제점이 발생한다. 이에 본 논문에서는 자원제공자의 성능과 논리적인 위치기반으로 자가 조직적 연산 오버레이 네트워크(Computation Overlay Network) 구성 기법과 자원제공자의 작업 완료 확률과 신용도 값을 이용하여 응용 QoS 보장을 위한 휴리스틱기반 복제 (Heuristic based Replication) 기법을 제안한다. 성능평가에서는 기존 스케줄링 기법과 자원제공자의 작업 완료 확률과 신용도에 따른 분포를 이용한 복제기반 스케줄링 기법을 비교평가 한다.

• International journal of advanced smart convergence
• /
• v.13 no.3
• /
• pp.143-149
• /
• 2024

As the size of data and models in machine learning training continues to grow, training on a single server is becoming increasingly challenging. Consequently, the importance of distributed machine learning, which distributes computational loads across multiple machines, is becoming more prominent. However, several unresolved issues remain regarding the performance enhancement of distributed machine learning, including communication overhead, inter-node synchronization challenges, data imbalance and bias, as well as resource management and scheduling. In this paper, we propose ParamHub, which utilizes orchestration to accelerate training speed. This system monitors the performance of each node after the first iteration and reallocates resources to slow nodes, thereby speeding up the training process. This approach ensures that resources are appropriately allocated to nodes in need, maximizing the overall efficiency of resource utilization and enabling all nodes to perform tasks uniformly, resulting in a faster training speed overall. Furthermore, this method enhances the system's scalability and flexibility, allowing for effective application in clusters of various sizes.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.38 no.4
• /
• pp.159-167
• /
• 2015

We consider a wafer lot transfer/release planning problem from semiconductor wafer fabrication facilities to probing facilities with the objective of minimizing the deviation of workload and total tardiness of customers' orders. Due to the complexity of the considered problem, we propose a two-level hierarchical production planning method for the lot transfer problem between two parallel facilities to obtain an executable production plan and schedule. In the higher level, the solution for the reduced mathematical model with Lagrangian relaxation method can be regarded as a coarse good lot transfer/release plan with daily time bucket, and discrete-event simulation is performed to obtain detailed lot processing schedules at the machines with a priority-rule-based scheduling method and the lot transfer/release plan is evaluated in the lower level. To evaluate the performance of the suggested planning method, we provide computational tests on the problems obtained from a set of real data and additional test scenarios in which the several levels of variations are added in the customers' demands. Results of computational tests showed that the proposed lot transfer/planning architecture generates executable plans within acceptable computational time in the real factories and the total tardiness of orders can be reduced more effectively by using more sophisticated lot transfer methods, such as considering the due date and ready times of lots associated the same order with the mathematical formulation. The proposed method may be implemented for the problem of job assignment in back-end process such as the assignment of chips to be tested from assembly facilities to final test facilities. Also, the proposed method can be improved by considering the sequence dependent setup in the probing facilities.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.9 no.5
• /
• pp.253-260
• /
• 2014

European Space Agency has recognized Integrated Modular Avionics and ARINC specification 653 as avionics computer system for space application. Integrated Modular Avionics specification reduces the space by integrating a system composed of many electronic devices into a computer. recent researches have been studying how to apply the ARINC 653 into an open source operating system, such as Linux. These studies have concentrated on partition scheduling for time separation. However, requirements to guarantee spatial separation should be further analyzed to ensure deterministic execution time. Therefore, memory management is needed to verify spatial isolation on Linux systems. This research proposes a new method to accomplish spatial isolation for the ARINC 653 specification in Linux. We have added new data structures and system calls to handle functionalities for spatial separation. They are used during the partition startup process. The proposed method was evaluated on the LEON4 processor, which is the next generation microprocessor to be used in the future space missions. All implementations confirm that spatial isolation of the ARINC 653 specification was accomplished.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.10
• /
• pp.4887-4907
• /
• 2017

Accurate traffic flow forecasting is critical to the development and implementation of city intelligent transportation systems. Therefore, it is one of the most important components in the research of urban traffic scheduling. However, traffic flow forecasting involves a rather complex nonlinear data pattern, particularly during workday peak periods, and a lot of research has shown that traffic flow data reveals a seasonal trend. This paper proposes a new traffic flow forecasting model that combines seasonal relevance vector regression with the hybrid chaotic simulated annealing method (SRVRCSA). Additionally, a numerical example of traffic flow data from The Transportation Data Research Laboratory is used to elucidate the forecasting performance of the proposed SRVRCSA model. The forecasting results indicate that the proposed model yields more accurate forecasting results than the seasonal auto regressive integrated moving average (SARIMA), the double seasonal Holt-Winters exponential smoothing (DSHWES), and the relevance vector regression with hybrid Chaotic Simulated Annealing method (RVRCSA) models. The forecasting performance of RVRCSA with different kernel functions is also studied.

• ETRI Journal
• /
• v.30 no.2
• /
• pp.183-193
• /
• 2008

In this paper, we propose quality of service mechanisms for flow-based routers which have to handle several million flows at wire speed in high-speed networks. Traffic management mechanisms are proposed for guaranteed traffic and non-guaranteed traffic separately, and then the effective harmonization of the two mechanisms is introduced for real networks in which both traffic types are mixed together. A simple non-work-conserving fair queuing algorithm is proposed for guaranteed traffic, and an adaptive flow-based random early drop algorithm is proposed for non-guaranteed traffic. Based on that basic architecture, we propose a dynamic traffic identification method to dynamically prioritize traffic according to the traffic characteristics of applications. In a high-speed router system, the dynamic traffic identification method could be a good alternative to deep packet inspection, which requires handling of the IP packet header and payload. Through numerical analysis, simulation, and a real system experiment, we demonstrate the performance of the proposed mechanisms.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.14 no.3
• /
• pp.186-195
• /
• 2006

In real-time control systems, the traditional timing analysis based on worst-case response-time(WCRT) is too conservative for the firm and soft real-time control systems, which permit the maximum utilization factor greater than one. We suggested a probabilistic analysis method possible to apply the firm and soft real-time control systems under considering dependency relationship between tasks. The proposed technique determines the deadline miss probability(DMP) of each task from computing the average response-time distribution under a fixed-priority scheduling policy. The method improves the predictable ability forthe average performance and the temporal behavior of real-time control systems.

• Journal of Korea Multimedia Society
• /
• v.14 no.7
• /
• pp.940-948
• /
• 2011

In this paper, we design and implement a task scheduler that considers real-time and fault tolerance in embedded system with a single processor. We propose a method how it can meet the deadlines of periodic tasks using RMS and complete the execution of aperiodic tasks by calculating surplus times from a periodic task set. And we describe a method how to recover of a transient fault task by managing backup time. We propose an important level of periodic tasks that can control the response time of periodic and aperiodic tasks. Finally, we analyse and evaluate the proposed methods by simulation.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2001.10a
• /
• pp.290-293
• /
• 2001

The mid-schedule planning of the ship is making a schedule about the process from cutting to erection. The ship consists of lots of blocks. This block has different process because of the shape of the block varies in accordance with the ship-type and the part position of the ship. The type and order of each block process initially must be generated for the mid-schedule planning. In this paper, the process planning, described as above is preparing the basic information before scheduler make a plan with the prepared manhour. The scheduling is done with this process planning which includes the information of the process order. This paper shows the research about three methods to design the process planning. First, investigate the expression method about information of the process planning for the mid-schedule planning in real workplace. Second, design the object of the process planning on the basis of investigating the expression method. Finally, develop the prototype of object on the basis of this designed process planning and then find the practical use in the mid-schedule planning. The object, which is developed in this paper, contains the main algorithm. In case of developing The Mid-Schedule Planning System, this object is expected to be utilized very easily as consisting another object.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.6
• /
• pp.1269-1275
• /
• 2013

Recently, many studies have suggested that an electric vehicle (EV) is one of the means for increasing the reliability of power systems in new energy environments. EVs can make a contribution to improving reliability by providing frequency regulation in power systems in which the Vehicle-to-Grid (V2G) technology has been implemented and, if economically viable, can be helpful in increasing power system reliability. This paper presents a stochastic method for optimal coordination of charging and frequency regulation decisions for an EV aggregator using the Least Square Monte-Carlo (LSMC) with modeling of electricity price uncertainty. The LSMC can be used to assess the value of options based on electricity price uncertainty in order to simultaneously optimize the scheduling of EV charging and regulation service for the EV aggregator. The results of a numerical example show that the proposed method can significantly improve the expected profits of an EV aggregator.