• Journal of Computing Science and Engineering
• /
• 제6권4호
• /
• pp.287-293
• /
• 2012

We introduce a methodology for the study of the application-level performance of time-sharing parallel jobs on a set of compute nodes in high performance clusters and report our findings. We assume that parallel jobs arriving at a cluster need to share a set of nodes with the jobs of other users, in that they must compete for processor time in a time-sharing manner and other limited resources such as memory and I/O in a space-sharing manner. Under the assumption, we developed a methodology to simulate job arrivals to a set of compute nodes, and gather and process performance data to calculate the percentage slowdown of parallel jobs. Our goal through this study is to identify a better combination of jobs that minimize performance degradations due to resource sharing and contention. Through our experiments, we found a couple of interesting behaviors for overlapped parallel jobs, which may be used to suggest alternative job allocation schemes aiming to reduce slowdowns that will inevitably result due to resource sharing on a high performance computing cluster. We suggest three job allocation strategies based on our empirical results and propose further studies of the results using a supercomputing facility at the San Diego Supercomputing Center.

• Journal of applied mathematics & informatics
• /
• 제15권1_2호
• /
• pp.511-525
• /
• 2004

we model communication and computer systems that process interactive and several and several types of background jobs. The scheduling policy in use is to share the processor among all interactive jobs and, at most, one background job of each type at a time according to the process sharing discipline. Background jobs of each type are served on a first-come-first-served basis. Such scheduling policy is called Processor Sharing with Background jobs (PSBJ). In fact, the PSBJ policy is commonly used on many communication and computer systems that allow interactive usage of the systems and process certain jobs in a background mode. In this paper, the stability conditions for the PSBJ policy are given and proved. Since an exact analysis of the policy seems to be very difficult, an approximate analytic model is proposed to obtain the average job sojourn times. The model requires the solution of a set of nonlinear equations, for which an iterative algorithm is given and its convergence is proved. Our results reveal that the model provides excellent estimates of average sojourn times for both interactive and background jobs with a few percent of errors in most of the cases considered.

• 한국경영과학회지
• /
• 제41권1호
• /
• pp.87-98
• /
• 2016

A single machine scheduling problem for jobs with stochastic processing time is considered in this study. Shortest processing time (SPT) sequencing according to the expected processing times of jobs is optimal for schedules with minimal expected mean flow time when all the jobs arrive to be scheduled and their expected processing times are known. However, SPT sequencing according to the expected processing time may not be optimal for the minimization of the mean flow time when the actual processing times of jobs are known. This study evaluates the complexity of SPT sequencing through a comparison of the mean flow times of schedules based on the expected processing times and actual processing times of randomly generated jobs. Evaluation results show that SPT sequencing according to the expected flow time exhibits chaotic variation to the optimal mean flow time. The relative deviation from the optimal mean flow time increases as the number of jobs, processing time, or coefficient of variation increases.

• 대한산업공학회지
• /
• 제35권4호
• /
• pp.257-265
• /
• 2009

This paper considers the hybrid flow shop scheduling problem for the objective of minimizing the number of tardy jobs. In hybrid flow shops, each job is processed through multiple production stages in series, each of which has multiple identical parallel machines. The problem is to determine the allocation of jobs to the parallel machines at each stage as well as the sequence of the jobs assigned to each machine. Due to the complexity of the problem, we suggest search heuristics, tabu search and simulated annealing algorithms with a new method to generate neighborhood solutions. In particular, to evaluate and select neighborhood solutions, three surrogate objectives are additionally suggested because not much difference in the number of tardy jobs can be found among the neighborhoods. To test the performances of the surrogate objective based search heuristics, computational experiments were performed on a number of test instances and the results show that the surrogate objective based search heuristics were better than the original ones. Also, they gave the optimal solutions for most small-size test instances.

• 한국화재소방학회논문지
• /
• 제27권3호
• /
• pp.85-93
• /
• 2013

본 연구에서는 소방대원이 수행하는 업무 중 중량물을 다루어야 하는 작업에 대한 분류 및 정량적 분석(NLE, 3DSSPP)을 수행하여 향후 소방대원의 중량물 관련 작업에 대한 위험도를 예측하고, 관련 위험도를 감소시키기 위한 기초 자료를 도출하고자 하였다. 도출된 위험도 등급은 위험도 점수가 3점부터 9점까지 총 7단계로 구성되었다. 가장 위험도가 높은 1등급은 나타나지 않았으며 인명구조 작업이 두 번째로 높은 2등급에 속하는 것으로 나타났다. 3등급은 파괴기구 활용, 환자 들어올리기작업 등으로 나타났다. 본 연구 결과들은 소방대원의 허리 부상을 포함하는 근골격계 질환 예방을 위한 작업방법 및 장비의 개선, 교육 프로그램의 개발에 기본 토대가 될 수 있다고 판단되며, 소방대원들의 근골격계 질환에 대한 지속적인 예방을 위해서는 본 연구의 골격을 토대로 한 추후 심층적인 조사와 연구가 시급하다고 사료된다.

• 대한교통학회지
• /
• 제36권3호
• /
• pp.195-202
• /
• 2018

직주균형은 일정한 지역에서 직장과 주거가 일치하는 상태를 의미한다. 이 연구는 직주균형이 교통안전에 미치는 영향을 분석하는 데 그 목적이 있으며, 이를 위해 광역권별 사고 모형 개발에 중점을 둔다. 주요 연구결과는 다음과 같다. 첫째, 통계적으로 유의한 총 3개의 일반화선형모형이 개발된다. 모든 모형에서 직주균형 요소는 교통사고 발생에 유의미한 영향을 미치는 것으로 판단된다. 둘째, 공통변수 중 주택보급률은 사고율 감소, 그리고 경제활동인구비율 및 출근통행 도착량은 사고율 증가에 영향을 미치는 것으로 분석된다. 따라서 모든 지역에서는 직주 불균형을 완화하려는 노력이 필요할 것으로 평가된다. 셋째, 수도권에서는 직주비와 업무통행비, 그리고 부산 울산광역권에서는 교차통근비에 중점을 둔 교통안전정책이 요구된다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권1호
• /
• pp.39-60
• /
• 2017

In the Cloud, for the scheduling of parallel jobs, there are many tasks in a job and those tasks are executed concurrently on different VMs (Visual machines), where each task of the job will be executed synchronously. The goal of scheduling is to reduce the execution time and to keep the fairness between jobs to prevent some jobs from waiting more time than others. We propose a Cloud model which has multiple Clouds, and under this model, jobs are in different lists according to the waiting time of the jobs and every job has different parallelism. At the same time, a new method-ZOMT (the scheduling parallel tasks based on ZERO-ONE scheduling with multiple targets) is proposed to solve the problem of scheduling parallel jobs in the Cloud. Simulations of ZOMT, AFCFS (Adapted First Come First Served), LJFS (Largest Job First Served) and Fair are executed to test the performance of those methods. Metrics about the waiting time, and response time are used to test the performance of ZOMT. The simulation results have shown that ZOMT not only reduces waiting time and response time, but also provides fairness to jobs.

• 한국임상보건과학회지
• /
• 제4권4호
• /
• pp.762-768
• /
• 2016

Purpose. This study was performed on survey related optician's perception of NCS and required jobs to develop curriculum of department of ophthalmic optics Methods. This study was to evaluate the questionnaire survey of 63 opticians working in Optometrist in Daegu Metropolitan City from November 1, 2016 to November 18, 2016. Results. As for the opticians who know about NCS, 23.8% of the respondents answered that the opticians' perception of NCS is very low. For the four required jobs of the technical part, less than 5 years of optometrists ; optometry 73.7% > fitting 47.4% > ophthalmic dispensing 36.8% > lensmeter 5.3%, more than 5 years and less than 10 years of optometrists ; optometry 84.6% > fitting 53.8% > lensmeter 46.2% > ophthalmic dispensing 38.5%, more than 10 years of optometrists ; optometry 67.7% > ophthalmic dispensing 51.4% > lensmeter 19.4% > fitting 16.1%. For the four required jobs of the management, less than 5 years of optometrists ; product 57.9% > service 47.4% > manners 10.6% > sales 5.3%, more than 5 years and less than 10 years of optometrists ; service 76.9% > product 53.9% > manners 46.2% > sales 38.5%, more than 10 years of optometrists ; service 45.2% > manners 42.0% > product 32.3% > sales 19.4%. Conclusions. Although there were 108 required jobs of the technical part and 94 required jobs of the management part, the technical level of the department of ophthalmic optics was satisfactory, but the classes of the management part were insufficient. It will be necessary to actively reflect the needs of industry through curriculums reform.

• 한국정보통신학회논문지
• /
• 제17권8호
• /
• pp.1934-1939
• /
• 2013

본 논문은 마감시간을 가진 작업들의 온라인 스케줄링 문제를 다룬다. 작업들이 시간이 지남에 따라 도착하고 스케줄링 알고리즘은 앞으로 도착할 작업들의 정보를 미리 알지 못한다. 작업들은 동일한 수행시간만큼 실행되고 알고리즘의 목표는 마감 시간 안에 수행을 완료한 작업들의 수를 최대화하는 것이다. 온라인 알고리즘의 성능은 모든 작업 정보를 미리 알고 최적의 답을 줄 수 있는 최적 알고리즘의 성능과 비교하는데 두 알고리즘 성능의 비를 경쟁비라고 한다. 일반적으로 정보의 부재로 인해 온라인 알고리즘은 큰 경쟁비를 가진다. 따라서 온라인 알고리즘에 보다 많은 머신 또는 보다 빠른 머신을 제공했을 때의 경쟁비를 계산하는 자원추가 분석을 수행할 수 있다. 본 논문에서는 온라인 알고리즘이 보다 많은 머신들을 이용할 수 있을 때 최적 알고리즘과 같은 성능을 낼 수 있음을 보일 것이다.

• 문화기술의 융합
• /
• 제9권6호
• /
• pp.181-186
• /
• 2023

본 연구에서는 청소년들의 4차 산업혁명의 인식 부족과 게임에 나타난 직업의 표현 양상 부족을 주장하고 게임을 통한 미래 유망 직업 직업교육을 제안한다. 다양한 선행연구들을 통해 4차 산업혁명에 따른 직업의 변화 청소년들의 직업교육 실태와 한계, 온라인게임의 교육적 효과 등을 검토하고 미래 유망 직업 관점에서 게임에 나타난 직업의 표현 양상 연구를 진행한 결과, 현재 모바일 시뮬레이션 게임 주인공 속 나타난 직업은 200개의 게임 중 45개에 불과했다. 미래 유망 직업의 직무를 살린 온라인게임의 양상과 청소년들의 희망 직업 순위 속 미래 유망 직업을 분석한 결과 미래 유망 직업은 '콘텐츠 크리에이터', '드론 전문가', '소프트웨어 개발자'가 표현되었다. 게임 주인공만의 직업을 조사했다는 한계를 지녔음에도, 현재 게임이 미래 유망 직업에 대한 직업교육의 기능적 측면을 충분히 포괄하고 있지 못함을 지적하고, 미래 유망 직업 반영의 필요성을 제시했다는 의의가 있다.