• Proceedings of the Korean Society of Computer Information Conference
• /
• 2018.07a
• /
• pp.15-16
• /
• 2018

본 논문에서는 다수의 센서 노드로 구성된 Internet of Things (IoT) 환경에서 새로운 환경에 대해 적응하는데 걸리는 시간을 줄이기 위한 새로운 스케줄링 기법을 제안한다. IoT 환경에서는 데이터 수집 및 전송 패턴이 사전에 정의되어 있지 않기 때문에 기존 정적인 Packet scheduling 기법으로는 한계가 있다. Q-learning은 네트워크 환경에 대한 사전지식 없이도 반복적 학습을 통해 Scheduling policy를 확립할 수 있다. 본 논문에서는 기존 Q-learning 스케줄링 기법을 기반으로 각 큐의 패킷 도착률에 대한 bound 값을 이용해 Q-table과 Reward table을 초기화 하는 새로운 Q-learning 스케줄링 기법을 제안한다. 시뮬레이션 결과 기존 기법에 비해 변화하는 패킷 도착률 및 서비스 요구조건에 적응하는데 걸리는 시간이 감소하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.9
• /
• pp.2213-2221
• /
• 2015

The ondemand governor used in android-based smartphone platforms is a CPU frequency scaling technique. The ondemand governor sets the CPU operating frequency depending on the CPU utilization rate. Job scheduling affects the CPU utilization rate. The power consumption is proportional to the value of operating frequency. Consequently, CPU frequency scaling and CPU utilization rate have an effect on power consumption in a smartphone. In this paper, we evaluated the performance of job scheduling techniques incorporating the ondemand governor in terms of CPU utilization, power consumption, and job deadline miss ratio.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.41 no.9
• /
• pp.681-688
• /
• 2013

Several problems caused by inevitable increment of airplane have been issued. The most effective solution to solve the issues is considered as establishing appropriate Air Traffic Management (ATM) that reduces aircraft's delay at an airport and intensify the airport's capacity. The purpose of this paper is to produce the optimum aircraft schedules that maximize the aircraft throughput by smooth air traffic flow near terminal area of an airport In this paper, mathematical formulations of the scheduling problem are firstly specified. Based on the mathematical modelling, an Optimization-Based Flight Scheduler that provides the optimum flight schedules for arriving aircraft is developed by introducing the Mixed Integer Linear Programming(MILP) and the Genetic Algorithms(GA). With this scheduler, we calculated the optimum schedules to compare to real schedule data from an Incheon Airport. As a result, it is validated that aircraft throughput produced by the optimum schedule is much better than that of the schedule from the Incheon airport. The optimization-based flight scheduler is expected to deal with problems due to the aircraft saturation in near future.

• The Journal of Korean Association of Computer Education
• /
• v.11 no.5
• /
• pp.67-76
• /
• 2008

Grid computing has a large scale virtual computing environment that enables a collaborative processing through sharing resources of geographically distributed organizations. In recent year, the development of wireless networks and mobile devices enables mobile devices to consider as a resource of the grids. However, there are some problems such as low performance of processors, small capacity of storages, limited capacity of battery, and low bandwidth. In this paper, to overcome these limitations occurred in mobile grid environments, we proposed a proxy-based mobile grid system. Our proposed system enables mobile devices to play roles as a resource consumer and a resource provider and to collaborate with wired grids through a mobile router. Also, we propose an adaptive job scheduling scheme to cope with context changes of mobile devices and compare our scheduling scheme with other scheduling schemes using a simulation tool, SimGrid, to verify the validity of our job scheduling scheme.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.1
• /
• pp.29-35
• /
• 2008

For efficient message passing of Parallel programs, it is required to schedule the involved two processes at the same time which are executed on different nodes, that is called 'co-scheduling' However, each node of cluster systems is built on top of general purpose multitasking OS. which autonomously manages local Processes. Thus it is not so easy to co-schedule two (or more) processes in such computing environment. Our work proposes a co-scheduling scheme for MPI-based parallel programs which exploits message exchange information between two parties. We implement the scheme on Linux cluster which requires slight kernel hacking and MPI library modification. The experiment with NPB parallel suite shows that our scheme results in 33-56% reduction in the execution time compared to the typical scheduling case. and especially better Performance in more communication-bound applications.

• The KIPS Transactions:PartA
• /
• v.16A no.6
• /
• pp.437-442
• /
• 2009

For multiprocessor systems, Earliest Deadline First (EDF) based on deadline and Least Laxity First (LLF) based on laxity are not suitable for practical environment since EDF has low schedulability and LLF has high context switching overhead. As a combining of EDF and LLF to improve the performance, Earliest Deadline Zero Laxity (EDZL) was proposed. EDZL is basically the same as EDF. But if the laxity of a task becomes zero, its priority is promoted to the highest level. In this paper, we propose Least Laxity Zero Laxity (LLZL) which is based on LLF. But context switching is allowed only if the laxity of a task on rady queue becomes zero. Simulation results show that LLZL has high schedulability approaching to LLF and low context switching overhead similar to EDF. In comparison with EDZL, LLZL has better performance in both of schedulability and context switching overhead.

• Journal of the Korea Society for Simulation
• /
• v.20 no.1
• /
• pp.19-28
• /
• 2011

The safety applications based on the IEEE 802.11p, periodically transmit the safety-related information to all surrounding vehicles with high reliability and a strict timeline. However, due to the high vehicle mobility, dynamic network topology and limited network resource, the fixed beacon scheduling scheme excess delay and packet loss due to the channel contention and network congestion. With this motivation, we propose a novel beacon scheduling algorithm referred to as spatial-aware(SA) beacon scheduling based on the spatial context information, dynamically rescheduling the beaconing rate like a TDMA channel access scheme. The proposed SA beacon scheduling algorithm was evaluated using different highway traffic scenarios with both a realistic channel model and 802.11p model in our simulation. The simulation results showed that the performance of our proposed algorithm was better than the fixed scheduling in terms of throughput, channel access delay, and channel load. Also, our proposed algorithm is satisfy the requirements of vehicular safety application.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.10
• /
• pp.367-372
• /
• 2016

In a modern smart building, sensors can detect various physical conditions, such as temperature, humidity, sound, motion, and light, which can be used in medical services and security, and for energy savings. This paper presents an efficient elevator scheduling system that utilizes smart sensor technologies with radio-frequency identification, video, and floor sensors to detect the arrival of elevator users in advance. The detected information is then delivered to the elevator scheduling system via building networks. By using this information, the proposed system makes a reservation call for efficient control of the elevator's direction and time. Experiments under a spectrum of traffic conditions show that the proposed system performs better than a legacy system with respect to average wait time, maximum wait time, and energy consumption.

• Journal of KIISE:Information Networking
• /
• v.33 no.3
• /
• pp.231-246
• /
• 2006

A server load balancer is used to accept and distribute client requests to one of servers in a wireless internet proxy server cluster. LVS(Linux Virtual Server), a software based server load balancer, can support several load balancing algorithms where client requests are distributed to servers in a round robin way, in a hashing-based way or in a way to assign first to the server with the least number of its concurrent connections to LVS. An improved load balancing algorithm to consider server performance was proposed where they check upper and lower limits of concurrent connection numbers to be allowed within each maximum server performance in advance and apply the static limits to load balancing. However, they do not apply run-time server load information dynamically to load balancing. In this paper, we propose a dynamic load balancing scheme where the load balancer keeps each server CPU load information at run time and assigns a new client request first to the server with the lowest load. Using a cluster consisting of 16 PCs, we performed experiments with static content(image and HTML). Compared to the existing schemes, experimental results show performance improvement in the cases of client requests requiring CPU-intensive processing and a cluster consisting of servers with difference performance.

• The KIPS Transactions:PartA
• /
• v.8A no.1
• /
• pp.16-26
• /
• 2001

This paper proposes an extended scheduler model that is an extension of the existing model proposed already in [4, 5], which consists of upper layer task scheduler and lower layer scheduling framework. However, in order to support aperiodic task scheduling, the task scheduler has been divided into two parts, such as periodic task control component and aperiodic task control component. Thus, the proposed model can support various bandwidth-preserving servers that can service aperiodic tasks. The model distinctly separates a classic monolithic kernel scheduler into several kernel components according to their functionality. This enables system developers to implement a new scheduling algorithm or aperiodic task server independent of complex low kernel mechanism, and reconfigure the system at need. In Real-Time Linux [6], we implemented the proposed scheduling framework representative scheduling algorithms, and server bandwidth-preserving servers on purpose to test. Throughout these implementations, we confirmed that a new algorithm or server could be developed independently without updates of complex low kernel modules. In order to verify efficiency of the proposed model, we measured the performance of several aperiodic task servers. The results showed this the performance of model, which even consisted of two hierarchical components and several modules, didnt have such high run-time overhead, and could efficiently support reconfiguration and scheduler development.