• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.7
• /
• pp.1524-1531
• /
• 2004

This study analyzes a load balancing technique using Round Robin Algorithm. The study uses two software packages (Packet Capture and Round Robin Test Package) to check packet volume from Virtual Network structure (data generator, virtual server, Server1,2,3), and finds out traffic distribution toward Server1,2 and 3. The functions of implemented Round Robin Load Balancing Monitoring System include round robin testing, system monitoring, and graphical indication of data transmission and packet volume. As the result of the study shows, Round Robin Algorithm allows servers to ensure definite traffic distribution, unless incoming data loads differ much. Although error levels are high in some cases, they were eventually alleviated by repeated tests for a long period of time.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10b
• /
• pp.1295-1298
• /
• 2001

멀티미디어 스트리밍 서비스를 제공하는 서버의 동적 부하분산을 위한 동적 가중치 기반 라운드 로빈 스케줄링 알고리즘을 제안한다. 기존의 가중치 기반 라운드로빈 알고리즘은 서버의 처리 용량만을 이용하여 가중치를 부여하므로 요청이 폭주할 경우 동적 부하 불균형을 갖게 된다. 동적 부하 불균형을 해결하기 위해 제안한 동적 가중치 기반 라운드로빈 알고리즘은 서버의 처리 용량뿐만 아니라 서버의 동적 부하를 이용하여 가중치를 부여하므로 동적 부하 불균형에 잘 적응하여 부하를 균형있게 조절한 수 있다. 제안한 알고리즘은 각 서버의 처리용량을 기준으로 가중치를 계산하고 동적으로 변하는 서버의 부하값에 가중치를 적용한다. 그 결과 동적 부하 불균형 문제를 해결했으며, 더 세밀한 부하 조절 기능을 수행할 수 있었다

• Journal of KIISE:Computer Systems and Theory
• /
• v.36 no.4
• /
• pp.291-303
• /
• 2009

Round robin based proportional share scheduling(RRPS) defines weight which determines share for each task and allocates CPU resource to each task in proportional to its respective weight. RRPS uses fairness as the measure of performance and aims at high fairness of scheduling. However, researches for scheduling fairness problem due to synchronization among tasks have been rarely investigated. In this paper, we discuss that scheduling delay due to synchronization may result high unfairness in RRPS. We explain such a situation as weight inversion. We then propose weight inheritance protocol(WIP), a synchronization mechanism, that prevents weight inversion. We also show that WIP can reduce unfairness using fairness analysis and simulation.

• Journal of Korea Multimedia Society
• /
• v.2 no.1
• /
• pp.22-29
• /
• 1999

In video-on-demand systems, movies are striped across the disks of the array in a video server to support multiple access. The blocks of the movies have to be placed carefully on the disks of the array in a video server in order to avoid load imbalance across these disks as well as to achieve maximum utilization of both space and bandwidth of these disks. This paper proposes a movie placement method called the GARR (group-alternation round-robin) that is based on the dynamic disk striping method considering the popularity of movie. The GARR uses group-alternation as movie selection policy and round-robin as data placement policy, where group-alternation allocates popular movie and unpopular movie groups to disks alternately. The performance of the GARR is evaluated by simulations.

• Journal of KIISE:Information Networking
• /
• v.32 no.6
• /
• pp.676-686
• /
• 2005

The Internet users as well as network providers are eager to have different qualities of service among users beyond the best-effort. In this paper, we propose a scheduling algorithm that provides a differentiated service in the granularity of user sessions. The proposed algorithm is a Hierarchical Deficit Round-Robin (H-DRR) algorithm that is an extension of an existing DRR algorithm. A main advantage is that H-DRR provides service differentiation for throughput-intensive applications such as FTP as well as delay-sensitive applications such as telnet or VoIP without distinguishing the types of applications. The most importance in providing a service differentiation in term of network providers is to have controllability and predictability. We show that H-DRR is superior to DRR in terms of controllability and predictability through both mathematical analysis and simulation experiments. Nevertheless, H-DRR requires O(1) complexity for implementation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.4
• /
• pp.846-853
• /
• 2004

This paper introduces implementation of a traffic distributed processing system on the Web Sever Networking. The study used two software packages （Packet Capture and Round-Robin Test Package） to check packet quantity from Virtual Network Structure (data generator, virtual server, Serve. 1, 2, 3), and could find out traffic distribution toward Server 1, 2, and 3. The functions of implemented Round-Robin Load Balancing Monitoring System include Round-Robin testing, system monitoring, and graphical indication of data transmission/packet quantity (figures ＆ diagram). As the result of the study shows, Round-Robin Algorithm ensured definite traffic distribution, unless incoming data loads differ much. Although error levels were high in some cases, they were eventually alleviated by repeated tests for a long period of time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.12C
• /
• pp.1207-1214
• /
• 2002

A scheduling scheme for high speed networks requires a low time complexity to schedule packets in a packet transmission time. High speed networks support a number of connections, different rates for each connection and variable packet length. Conventional round robin algorithms have a time complexity of O(1), but their short time fairness, latency and burstiness depend on the quantum of a connection due to serving several packets for a backlogged connection once a round. To improve these properties, we propose in this paper an efficient packet scheduling scheme which is based on the credits of a connection and has a time complexity of O(1). We also analyzed its performance in terms of short time fairness, latency and burstiness. The analysis results show that the proposed scheme can improve the performance compared with traditional round robin schemes. The proposed scheme can be easily utilized in high speed packet networks.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.9A
• /
• pp.973-982
• /
• 2004

High-speed scheduling algorithms are required for high-performance input-queued switches to achieve good performance. Various Round-Robin scheduling algonthms for Virtual-Output-Queue (VOQ) switch architectures have been proposed, hke iSLIP, DRRM (Dual Round-Robin Matching). iSLIP can achieve high performance and have already been implemented in hardware. DRRM has been proved to achieve better performance and simpler than iSLIP But neither iSLIP nor DRRM can efficiently solve the problem of the Round-Robm pointers' desynchronization. In this paper, we have proposed "Well-Regulated Pseudo-request Dual Round-Robin Matching" Algorithm. It is developed from DRRM, and can always keep the pointers' desynchronization. Since our algorithm is based on the Round-Robin scheduling, it is also simple to be implemented. And simulation results also show that our proposed algonthm performs pretty well under various, traffic models.

• Proceedings of the Korea Contents Association Conference
• /
• 2013.05a
• /
• pp.33-34
• /
• 2013

범용 운영체제 윈도우즈는 라운드로빈(round-robin)스케줄링 방식을 사용하고 있다. 라운드로빈 스케줄링 방식은 높은 우선순위 태스크에 대한 빠른 응답성을 보장할 수 없어 실시간성을 제공하지 못한다. 이를 해결하기 위해 윈도우즈에서 실시간성을 제공하기 위한 RTiK-MP(Real Time implant Kernel-Multi Processor)가 개발되었다. 기존 RTiK-MP는 멀티프로세서 윈도우즈 환경에서 정상동작하게 되어있으나 서비스팩(Service Pack)환경에서는 Local APIC 타이머 레지스터의 값이 초기화되면서 윈도우즈에 실시간성을 제공하지 못하는 문제점이 있다. 본 논문에서는 윈도우즈 서비스팩에서 발생하는 RTiK-MP의 문제점에 대해 설명하고 이에 대한 해결방안을 기술하여 윈도우즈 서비스팩에서 실시간성을 제공하기 위한 방안을 연구하였다.

• The Journal of the Korea Contents Association
• /
• v.2 no.4
• /
• pp.85-92
• /
• 2002

This paper proposes the cell scheduling algorithm proper to high-speed ATM switch. The proposed algorithm is the VRR(Variable weight Round Robin) combined the FRR to variant QLT. The FRR performs the cell service of the fixed weight by each buffer. So, FRR don't support the QoS of ATM service classes although it is easy to implement a high speed switch. VRR uses the method expaned to variable weight according to buffer state as well as schedules the cell according the Fixed weight based FRR. The simulation to evaluate the proposed algorithm was done by AweSim arid Visual C++. The result graphs show that the proposed algorithm is excellent, especially in the aspect of cell loss. This area is engaged by English Abstract.