• Journal of Korea Society of Industrial Information Systems
• /
• v.17 no.5
• /
• pp.29-36
• /
• 2012

Cloud computing has gained great popularity because users don't need to install any softwares as well as maintain hardwares and service providers also can utilize its resources through virtualization of servers. As workflows feature variableness and servers are heterogenous, efficient scheduling of workflows in cloud computing is important factor in view of responsibility and resource utilization. In this paper, we propose a new workflow scheduling named 2-step scheduling which prioritizes each workflow through its significance degree and allocates resources to workflows through decision table. The goal of this 2-step scheduling is to improve responsibility as well as availability versus cost. Simulation results show that the proposed scheme in contrast of counterparts can improve the responsibility as well as availability of resource.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.11
• /
• pp.1701-1713
• /
• 1993

This paper proposes a simple and highly effective RACMC(Real-time ATM Cell Monitoring and Control) algorithm and the resulting bandwidth gain effects art considered. RACMC algorithm performs usage parameter controls according to the monitoring informations of input data cells generated from the accepted connections and the controlling informations set by the M/P(Management Plane) for that connection. The results of monitoring and controlling actions for ATM data cells are transmitted to the M/P and the control parameters in lookup table are updated according to the condition of currently used bandwith. Therefore, the proposed algorithm can allocate network resource optimally and solve the several tantalizing problems that the existing cell control algorithm have, that is, the difficulty in controlling as monitoring very bursty traffics, unavoidable processing delay, and limited input buffer size when implemented. By the performance analysis using computer simulation, RACMC algorithm proves to be very effective especially in ATM network as implemented simply.

• Journal of Korea Society of Industrial Information Systems
• /
• v.8 no.3
• /
• pp.19-31
• /
• 2003

For the implementation of large scale GRID systems, the performance scalability in resource scheduling is clearly to be addressed. In this research, we analyzed existing scheduling frameworks from the viewpoint of the performance and propose a novel resource scheduling framework called resource parameter based scheduling. Proposed scheduling framework consists of three components. The first is the resource parameter graph that expresses resource information via inter-resource relation and the composition base on the hierarchical structure. The second component is the resource parameter tree to be used for the implementation of the memory-based index of resource information. The third component is the resource information repository which mostly consists of static data to be used for the general resource information services. This paper presents the details of the framework.

• Proceedings of the IEEK Conference
• /
• 2001.06b
• /
• pp.121-124
• /
• 2001

This paper proposed resource allocation algorithm for the minimum switching activity of functional unit in high level synthesis process as like DSP which is circuit to give many functional unit. The resource allocation method after scheduling use the power function calculating average hamming distance and switching activity of the between two input. First of all, the switching activity is calculated by the input value after calculating the average hamming distance between operation. In this paper, the proposed method though high If level simulation find switching activity in circuit each functional unit exchange for binary sequence length and value bit are logic one value. To use the switching activity find the allocation with minimal power consumption, the proposed method visits all control steps one by one and determines the allocation with minimal power consumption at each control step. As the existing method, the execution time can be fast according to use the number of operator and max control step. And it is the reduction effect from 6% to 8%.

• Journal of Korean Institute of Industrial Engineers
• /
• v.23 no.2
• /
• pp.433-450
• /
• 1997

In this study, a heuristic algorithm is developed to solve a resource-constrained scheduling problem. The problem involves multiple projects and multiple resource categories, and allows flexible resource allocation to each activity. The objective is to minimize the maximum completion time. The algorithm takes advantage of the basic structure of a heuristic algorithm, called the exchange heuristic, but employs different strategies on some critical steps of the original algorithm which have significant effects on the algorithm performance. The original algorithm and the modified algorithm were compared through an experimental investigation. The modified algorithm produces significantly shorter schedules than the original algorithm, though it requires up to three times more computation time.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.05a
• /
• pp.994-997
• /
• 2013

가비지 컬렉션(Garbage Collection)은 시스템에서 더 이상 사용하지 않는 동적 할당된 메모리 블록 혹은 개체를 찾아 자동적으로 다시 사용 가능한 자원으로 회수하는 것을 의미한다. 최근 대부분의 프로그래밍 언어에서는 메모리 관리를 자동으로 처리해주는 가비지 컬렉터를 기본적으로 포함하고 있으며 이러한 시스템 환경은 개발자들의 개발 속도 향상과 프로그램 가독성을 높여주는 이점을 주고 있다. 그러나 가비지 컬렉터는 자원이 한정되어 있는 스마트폰과 같은 환경에서는 큰 오버헤드를 가지며 성능 저하의 주 원인으로 꼽히기도 한다. 따라서 iOS의 경우에는 가비지 컬렉터를 지원하지 않는다. 이에 따라 본 연구에서는 스마트폰의 안드로이드와 iOS의 프로그래밍 언어인 Java와 Object C의 가비지 컬렉터의 알고리즘을 분석하여 두 언어의 개발환경의 차이를 비교 하였다. 또한 앞으로 Object C의 메모리 관리 정책에 대하여 서술하였다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.6
• /
• pp.187-193
• /
• 2015

The ICT healing platform sets a couple of goals including preventing chronic diseases and sending out early disease warnings based on personal information such as bio-signals and life habits. The 2-step open system(TOS) had a relay designed between the healing platform and the storage of personal health data. It also took into account a publish/subscribe(pub/sub) service based on large-scale connections to transmit(monitor) the data processing process in real time. In the early design of TOS pub/sub, however, the same buffers were allocated regardless of connection idling and type of message in order to encode connection messages into a deflate algorithm. Proposed in this study, the dynamic buffer allocation was performed as follows: the message transmission type of each connection was first put to queuing; each queue was extracted for its feature, computed, and converted into vector through tf-idf, then being entered into a k-means cluster and forming a cluster; connections categorized under a certain cluster would re-allocate the resources according to the resource table of the cluster; the centroid of each cluster would select a queuing pattern to represent the cluster in advance and present it as a resource reference table(encoding efficiency by the buffer sizes); and the proposed design would perform trade-off between the calculation resources and the network bandwidth for cluster and feature calculations to efficiently allocate the encoding buffer resources of TOS to the network connections, thus contributing to the increased tps(number of real-time data processing and monitoring connections per unit hour) of TOS.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.44-46
• /
• 2017

본 논문에서는 클라우드 컴퓨팅에서 서버통합과 동시에 서버들의 자원 활용률을 증가시키기 위해 실행되는 가상머신 재배치가 수행되고 난 후에 가상머신들의 상태를 관리하기 위한 기법에 관한 모델을 제안한다. 가상머신 재배치에는 많은 문제점이 발생한다. 재배치 알고리즘으로 인한 서버 통합 시간이 증가되는 문제점도 있지만 재배치를 수행하면서 발생하는 가상머신 마이그레이션으로 인한 전체적인 클라우드 시스템의 부하를 예로 들 수 있다. 제안하는 기법은 가상머신들의 재배치 알고리즘 수행 도중 발생 할 수 있는 문제점과 재배치 이후에 가상머신들의 구동 시간이 종료되어 자연스럽게 할당이 해제되는 경우에 발생하는 문제점을 해결 할 수 있다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.12
• /
• pp.981-987
• /
• 2000

이 논문은 상위 단계 데이터-경로 합성에서 연산 스케쥴링과 자원 할당 및 배정을 동시에 고려한 통합적 접근 방법을 제시한다. 제안한 방법은 스케쥴링 되어있지 않은 데이터-플로우 그래프에 대해서 수행에 필요한 총 clock 스텝 수와 필요한 회로 면적을 동시에 최소화하는 데이터-경로 생성에 특징이 있다. 일반적으로, 연결선의 결정이 합성의 마지막 단계에서 이루어지는 기존의 방법과는 다르게, 우리의 접근 방법은 연산 스케쥴링과 연산의 연산 모듈 배정 그리고 변수의 레지스터 배정 작업을 동시에 수행하여 추가적인 연결선의 수를 매 clock 스텝마다 최적화(optimal) 시킨다. 본 논문은, 이 문제를 최소-비용의 최대-플로우 문제로 변형하여 minimum cost augmentation 방법으로 polynomial time 안에 해결하는 알고리즘을 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10b
• /
• pp.1103-1106
• /
• 2000

정체(congestion)가 발생한 네트워크에서 자원관리 시스템은 정체를 제어하고 프레임 상실(loss)을 최소화함으로써 리얼타임 트래픽에 대한 요구를 충족시키기 위해 절대적으로 요구된다. 그에 따라 본 논문에서는 외부출력 인터페이스(Outgoing Interface)를 통과한 각종 멀티미디어 프레임들의 정체나 상실을 최소화하는 것을 기본 목적으로 하여 임계작업(mission critical)의 애플리케이션이 요청한 트래픽이 삽입되는 큐에 프레임이 존재하는 한, 부 스케줄러(Child Scheduler)를 생성하여 연속적으로 처리를 해주는 한편, 다른 트래픽들은 주 스케줄러(Parent Scheduler)가 할당된 대역폭만큼 Round-Robin 방식으로 계속 처리를 해주도록 하여 기아(starvation)가 일어나지 않도록 하는 CQ(Custom Queuing) 기반의 대역폭 공유 큐잉 알고리즘을 설계하였다.