• 지구물리
• /
• 제4권3호
• /
• pp.163-173
• /
• 2001

This paper deals with scheduling problems which are the most important subtask in High-Level Synthesis. Especially, we have concentrated our attentions on the data-path scheduling which can get the structural informaion trom the behavioral algorithm as a first step in synthesis procedure. Suggest Forward scheduling methode is executed the ASAP and ALAP scheduling to use the fifth - order elliptic wave filter of a standard benchmark moedl, and then it is drawing up T.N matrix table by the number resource and control-step, using the table extract of the simple than down-limit value of the control-step for the number of given resource th use this table. All of existing list scheduling techniques determine the priority function first, and then do the operation scheduling, But, the suggested forward scheduling technique does the schedule first, and determines the priority functions if needed in scheduling process.

• 공학논문집
• /
• 제4권1호
• /
• pp.35-45
• /
• 2002

This paper deals with the scheduling problems which are the most important subtask in High-Level Synthesis. Especially, we have concentrated our attentions on the data-path scheduling which can get the structural informations from the behavioral algorithm as a first step in synthesis procedure. Suggest Forward scheduling methode is executed the ASAP and ALAP scheduling to use the fifth – order elliptic wave filter of a standard benchmark model, and then it is drawing up T.N matrix table by the number of resource and control-step, using the table extract of the simple than down-limit value of the control-step for the number of given resource to use this table. All of existing list scheduling techniques determine the priority functions first, and then do the operation scheduling, But, the suggested forward scheduling technique does the schedule first, and determines the priority functions if needed in scheduling process.

• 지능정보연구
• /
• 제13권3호
• /
• pp.63-82
• /
• 2007

본 논문에서는 유연제조시스템(FMS)에서 다단계스케줄링 문제를 효율적으로 해결하기 위한 적응형 혼합유전 알고리즘(ahGA) 접근법을 제안한다. 제안된 ahGA는 FMS의 해를 개선시키기 위하여 이웃탐색기법을 사용하며, 유전탐색과정에서의 수행도를 향상시키기 위해 유전알고리즘(GA)의 파라메터들을 조정하기 위한 적응형 구조를 사용한다. 수치실험에서는 제안된 ahGA와 기존의 알고리즘들 간의 수행도를 비교하기 위하여 두가지형태의 다단계스케줄링문제를 제시한다. 실험결과는 제안된 ahGA가 기존의 알고리즘들 보나 더 뛰어난 수행도를 보여주고 있다.

• 한국인터넷방송통신학회논문지
• /
• 제24권3호
• /
• pp.63-67
• /
• 2024

최근 스마트 물류, 핀테크, 엔터테인먼트 등 다양한 산업 분야의 인공지능 워크로드들이 클라우드 상에서 실행되고 있다. 본 논문은 이기종 GPU 클러스터로 구성된 다중 테넌트 클라우드 시스템에서 다양한 인공지능 워크로드가 실행될 때 발생하는 스케줄링 문제를 다룬다. 전통적인 스케줄링은 이러한 환경에서 GPU 이용률을 크게 저하시켜 시스템의 성능을 떨어뜨린다. 이러한 문제를 해결하기 위해, 본 논문에서는 유전 알고리즘 기반의 최적화 기법을 사용하는 새로운 스케줄링 접근 방식을 제안하고, 이를 프로세스 기반 이벤트 시뮬레이션 프레임워크에 구현하였다. 알리바바의 MLaaS 클러스터에서 수집한 광범위한 인공지능 작업들의 트레이스를 재현하는 실험을 통해 제안하는 스케줄링이 기존 스케줄링에 비해 GPU 이용률을 크게 개선함을 확인하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -1
• /
• pp.212-215
• /
• 2002

In this paper, we present an intermediate representation CDFG(Control Data Flow Graph) and an efficient scheduling technique for low power circuit design. The proposed CDFG represents control flow, data dependency and such constraints as resource constraints and timing constraints. In the scheduling technique, the constraints are substituted by subgraphs, and then the number of subgraphs is minimized by using the inclusion and overlap relation efficiently. Also, iterative rescheduling process are performed in a minimum bound estimation, starting with the as soon as possible as scheduling result, so as to reduce the power consumption in low power design. The effectiveness of the proposed algorithm has been proven by the experiment with the benchmark examples.

• 대한산업공학회지
• /
• 제23권4호
• /
• pp.621-633
• /
• 1997

This paper is concerned with the multiobjective vehicle scheduling problem with time and area-dependent travel speeds(MVSPTD), in which two conflicting objectives are explicitly treated and the travel speed between two locations depends on the passing area and time of day. The two objectives are the minimization of total vehicle travel time and the minimization of total weighted tardiness. First, I construct a mixed integer linear programming formulation of the MVSPTD, and present o heuristic algorithm that builds the vehicle schedules based on the savings computed. The results of computational experiments showed that the heuristic performs very well. Finally, I propose an expert system for vehicle scheduling in the MVSPTD. Its whole process is executed under VP-Expert expert system environment.

• ETRI Journal
• /
• 제32권5호
• /
• pp.657-664
• /
• 2010

Many embedded multimedia systems employ special hardware blocks to co-process with the main processor. Even though an efficient handling of such hardware blocks is critical on the overall performance of real-time multimedia systems, traditional real-time scheduling techniques cannot afford to guarantee a high quality of multimedia playbacks with neither delay nor jerking. This paper presents a hardware-aware rate monotonic scheduling (HA-RMS) algorithm to manage hardware tasks efficiently and handle special hardware blocks in the embedded multimedia system. The HA-RMS prioritizes the hardware tasks over software tasks not only to increase the hardware utilization of the system but also to reduce the output jitter of multimedia applications, which results in reducing the overall response time.

• 한국통신학회논문지
• /
• 제5권1호
• /
• pp.83-87
• /
• 1980

階層構造 COMPUTER Network을 通한 工程制御로 Processing을 分數하여 Fault tolerance를 極大化시키며 複雜하고 多樣한 變數의 相互關係를 週期的으로 監視制御하는 分散制御 Processor JOB은 그 週期와 實行時間으로 定義할 수 있다. 모든 JOB에 대하여 Tree structure인 關係를 가진 subset들로 구성하여 이에 JOB Scheduling Algorithm을 求하여 본 결과 FCFS(Fist Come/First Service)인 Schedule 보다 Processor의 利用에 있어 Loose Time을 減少시키고 處理 可能時間 確保에 有利하였다.

• 한국항해학회지
• /
• 제24권5호
• /
• pp.361-371
• /
• 2000

This paper treats a genetic algorithm for ship scheduling problem in set packing formulation. We newly devised a partition based representation of solution and compose initial population using a domain knowledge of problem which results in saving calculation cost. We established replacement strategy which makes each individual not to degenerate during evolutionary process and applied adaptive mutate operator to improve feasibility of individual. If offspring is feasible then an improve operator is applied to increase objective value without loss of feasibility. A computational experiment was carried out with real data and showed a useful result for a large size real world problem.

• 산업경영시스템학회지
• /
• 제42권1호
• /
• pp.33-40
• /
• 2019

This study focuses on a job-shop scheduling problem with the objective of minimizing total tardiness for the job orders that have different due dates and different process flows. We suggest the dispatching rule based scheduling algorithm to generate fast and efficient schedule. First, we show the delay schedule can be optimal for total tardiness measure in some cases. Based on this observation, we expand search space for selecting the job operation to explore the delay schedules. That means, not only all job operations waiting for process but also job operations not arrived at the machine yet are considered to be scheduled when a machine is available and it is need decision for the next operation to be processed. Assuming each job operation is assigned to the available machine, the expected total tardiness is estimated, and the job operation with the minimum expected total tardiness is selected to be processed in the machine. If this job is being processed in the other machine, then machine should wait until the job arrives at the machine. Simulation experiments are carried out to test the suggested algorithm and compare with the results of other well-known dispatching rules such as EDD, ATC and COVERT, etc. Results show that the proposed algorithm, MET, works better in terms of total tardiness of orders than existing rules without increasing the number of tardy jobs.