• 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.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.11S
• /
• pp.3684-3693
• /
• 2000

In this paper, performance of a nonblocking high speed packet switch with switching capacity m which adopts a dynamic priority control function (DPCF) of a threshold- based Bernoulli priority jump (TBPJ) scheme is considered. Each input queue has two separate buffers with different sizes for two classes of traffics, delay-sensitive and loss-sensitive traffics, and adopts a TBPJ scheme that is a general state-dependent Bernoulli scheduling scheme. Under the TBP] scheme, a head packet of the delay-sensitive traffic buffer goes into the loss -sensitive traffic buffer with Hernoulli probability p according to systems states that represent the buffer thresholds and the number of packets waiting for scheduling. Performance analysis shows that TBPJ scheme obtains large performance build-up for the delay-sensitive traffic without performance degradation for the loss-sensitive traffic. In addition to, TBP] scheme shows better performance than that of HOL scheme.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.5 no.9
• /
• pp.1492-1512
• /
• 2011

Due to multiple hops, mobility and time-varying channel, supporting delay sensitive real-time traffic in wireless local area network-based (WLAN) mesh networks is a challenging task. In particular for real-time traffic subject to medium access control (MAC) layer control overhead, such as preamble, carrier sense waiting time and the random backoff period, the performance of real-time flows will be degraded greatly. In order to support real-time traffic, an efficient adaptive packet scheduling (APS) scheme is proposed, which aims to improve the system performance by guaranteeing inter-class, intra-class service differentiation and adaptively adjusting the packet length. APS classifies incoming packets by the IEEE 802.11e access class and then queued into a suitable buffer queue. APS employs strict priority service discipline for resource allocation among different service classes to achieve inter-class fairness. By estimating the received signal to interference plus noise ratio (SINR) per bit and current link condition, APS is able to calculate the optimized packet length with bi-dimensional markov MAC model to improve system performance. To achieve the fairness of intra-class, APS also takes maximum tolerable packet delay, transmission requests, and average allocation transmission into consideration to allocate transmission opportunity to the corresponding traffic. Detailed simulation results and comparison with IEEE 802.11e enhanced distributed channel access (EDCA) scheme show that the proposed APS scheme is able to effectively provide inter-class and intra-class differentiate services and improve QoS for real-time traffic in terms of throughput, end-to-end delay, packet loss rate and fairness.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.39 no.4
• /
• pp.188-200
• /
• 2002

The WRR scheduling algorithm is widely used in ATM networks due to its simplicity and the low cost of hardware implementation. It guarantees minimum cell rate according to the weight of each queue. The fairness is a important factor for ABR service. That is, scheduling algorithm allocates network resources fairly to each VC. However, WRR algorithm shows worse performance on bursty traffic. Because it schedules input traffics according to predetermined weight, it can not satisfies fairness criteria, MCR plus equal share and Maximum of MCR or Max-Min share, defined by ATM Forum TM 4.1 specification. The Nabeshima et al algorithm is not adapt to network status rapidly because it is not compensate the weights of unused bandwidth to VCs and assign the unused bandwidth to VCs by RR method. In this paper, we propose a scheduling algorithm for satisfying the two fairness criteria, MCR plus equal share and Maximum of MCR or Max-Min share, among the six criteria defined by ATM Forum TM 4.1 specification. The WRR, Nabeshima et al, and the proposed scheduling algorithms are compared with respect to fairness and convergence time throughout experimental simulation. According to the simulation results, the proposed algorithm shows higher fairness and more rapid convergence than other algorithms.

• Journal of Korea Multimedia Society
• /
• v.13 no.9
• /
• pp.1285-1298
• /
• 2010

Since the TinyOS incorporating a non-preemptive task scheduling policy uses a FIFO (First-In First-Out) queue, a task with the highest priority cannot preempt a task with lower priority before the task with lower priority must run to completion. Therefore, the non-preemptive TinyOS cannot guarantee the completion of real-time user tasks within their deadlines. Additionally, the non-preemptive TinyOS needs to meet the deadlines of user tasks as well as those of TinyOS platform tasks called by user tasks in order to guarantee the deadlines of the real-time services requested by user tasks. In this paper, we present a group-based real-time scheduling technique that makes it possible to guarantee the deadlines of real-time user tasks in the TinyOS incorporating a non-preemptive task scheduling policy. The proposed technique groups together a given user task and TinyOS platform tasks called and activated by the user task, and then schedule them as a virtual big task. A case study shows that the proposed technique yields efficient performance in terms of guaranteeing the completion of user tasks within their deadlines and aiming to provide them with good average response time, while maintaining the compatibility of the existing non-preemptive TinyOS platform.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11a
• /
• pp.562-564
• /
• 2005

최근 지역적으로 분산되어 있는 이질적인 고성능 컴퓨팅 자원을 하나로 묶어 거대한 시스템을 구성하는 그리드 컴퓨팅(Grid Computing)에 대한 연구가 활발하게 이루어짐에 따라, 본 논문에서는 그리드 시스템에서 클라이언트 작업들을 특성에 따라 분류하여 우선순위가 높은 작업은 예약이 가능한 큐로 분배하고 우선순위가 낮은 작업은 백필이 가능한 큐로 할당하는 다중 큐 스케줄링 기법을 제안하였고 다양한 실험을 통하여 제안된 기법들의 성능을 평가하였다. 그 결과 그리드 컴퓨팅 시스템의 이용률(Utilization)이 높아지고, 작업 지연시간(Slowdown)이 줄어드는 것을 확인하였다.

• ETRI Journal
• /
• v.35 no.1
• /
• pp.18-26
• /
• 2013

This paper proposes an efficient bandwidth utilization (EBU) algorithm that utilizes the unused bandwidth in dynamic bandwidth allocation (DBA) of a 10-gigabit-capable passive optical network (XGPON). In EBU, an available byte counter of a queue can be negative and the unused remainder of an available byte counter can be utilized by the other queues. In addition, EBU uses a novel polling scheme to collect the requests of queues as soon as possible. We show through analysis and simulations that EBU improves performance compared to that achieved with existing methods. In addition, we describe the hardware implementation of EBU. Finally we show the test results of the hardware implementation of EBU.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.05a
• /
• pp.757-760
• /
• 2006

고성능 컴퓨팅 자원(Computing Resources)을 네트워크를 통하여 하나로 연동하여 단일 컴퓨팅 작업을 수행하는 그리드 컴퓨팅(Grid Computing)에 대한 연구가 최근 활성화 되고 있다. 이에 본 논문에서는 그리드를 구성하는 노드에 작업을 분배하는 메타 스케줄링(Metascheduling)과 노드내의 작업을 프로세서에 적절히 할당하는 작업 스케줄링 방법을, 작업 큐와 두 개의 백필 큐를 사용하는 혼합 기법을 사용하여, 작업 지연율(Slowdown Ratio)을 줄이고 그리드 컴퓨팅의 이용률(Utilization)을 높이는 방법을 제안하였으며 다양한 실험을 통하여 성능을 평가하였다.

• The KIPS Transactions:PartA
• /
• v.11A no.7 s.91
• /
• pp.577-582
• /
• 2004

Double-ended Proirity queues(DEPQ) can be used in applications such as scheduling or sorting. The data structures for DEPQ can be con-structed with or without pointers. The implicit representation without pointers uses less memory space than pointer-based representation. This paper presents a novel fast implicit heap called 4-deapr$\ast$ which utilizes cache memory efficiently. Experimental results show that the 4-deap$\ast$ is faster than symmetric min-max heap as well as deap.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.12B
• /
• pp.2003-2015
• /
• 2000

본 논문에서는 ATM 망을 구성하고 있는 노드(node) 간의 셀(cell) 전송시에 출력 링크(output link)로 전송된 셀이 수신 노드에서 수행하는 UPC/NPC에 의해 손실되지 않는 트래픽 정형화기(shaper)를 제안한다. 제안한 정형화기는 비실시간 가변비트율(nrt-VBR) 트래픽을 대상으로 한 것으로 각 ATM 연결(connection)에 대해 독립된 큐(queue)를 할당하였으며, 출력 링크로 전송할 셀을 저장하고 있는 큐를 선정하기 위해 모든 큐에 대해 Generic Cell Rate Algorithm(GCRA)를 수행하여 큐의 상태를 결정하고, 그 결과 전송 가능 상태에 있는 모든 큐들을 대상으로 본 논문에서 제안한 시간의존 확률적 스케줄링(time-dependent probability scheduling) 알고리즘을 적용하여 특정 큐를 선택하여 셀을 전송한다. 제안한 정형화기의 성능평가를 위해 WRR(Weighted Round Robin) 스케줄링 알고리즘을 사용한 정형화기 및 WFQ(Weighted Fair Queueing) 스케줄링 알고리즘을 적용한 정형화기에 대한 모의실험을 하였으며, 그 결과 제안한 정형화기는 비실시간 가변비트율(nrt-VBR) 트래픽의 정형화(shaping)에 우수한 성능을 나타냄을 확인하였다.