寬頻交換機的高效能存取控制及頻寬分配

Abstract

寬頻非同步傳輸模式╱同步傳輸模式（ATM/STM）交換網路，被期望使 用可用的網路資源，提供多媒體的服務。其中，關鍵性的資源是ATM/STM 交換機。為了提供網路訊務（traffic）品質服務（QoS），控制交換機的 存取是非常重要的。為了這個目標，我們針對ATM及STM交換機，分別發展 高效能存取控制和頻寬分配。關於ATM交換機的存取控制，我們針對輸入 緩衝（input-queued）多對一（multicast）ATM交換機，提出一個能達成 高系統效能、低分封（packet）延遲及低分封遺失機率的輸入存取方法。 每個輸入埠的多對一及一對一分封被分開存放。多對一緩衝器比一對一緩 衝器有較高的優先權，而且，兩類緩衝器皆以循環優先權（cyclic- priority）的方式來達成公平服務。特別地，每個一對一緩衝器是以視窗 服務（window-service）的基礎來處理，每個多對一緩衝器則是以一次射 出（one-shot）的方式來交換。為了評估此存取方法，我們以簡略的循環 優先權模式為基礎，對單一頻道（unichannel）及多重頻道（ multichannel）的交換機，提出近似性的分析。我們假設到達交換機輸入 埠的一對一及多對一分封皆具有白努力（Bernoulli）過程的特性，並且 ，把視窗服務及一次射出規劃皆列入考慮。我們也藉由模擬結果，來顯示 近似性分析的正確性，並以系統效能、平均分封延遲及分封遺失機率，來 驗證我們所提方法的優越性。至於STM交換機的頻寬分配，我們則著重在 將頻寬分配給聲音及資料訊務。我們的目標是，藉由具有非同類（ heterogeneous）到達訊務及多重特定頻道的佇列模式（queueing ），決 定分配給聲音及資料訊務的最佳頻寬。而分析結果，則是以模擬結果來驗 證其正確性。藉由分析，我們提出了一個多項有限的（polynomial- bounded）演算法來建立頻寬分配範例。利用頻寬分配範例，網路頻寬便 能有效的分配給聲音及資料訊務，並保證最小的資料延遲及聲音阻塞機率 。因此，在不同的網路負載量下，頻寬分配範例確保了資料延遲及聲音阻 塞機率的品質服務。 總之，我們可以將輸入存取方法與頻寬分配範例 ，以整合的、有效的方式結合在一起，藉以改善網路效能並滿足不同品質 服務的要求。 Broadband Asynchronous Transfer Mode / Synchronous Transfer Mode (ATM/STM) switching networks are expected to provide multimedia services using available network resources. One of the critical resources is the ATM/STM switching. To support the offered traffic at an acceptable Quality-of-Service (QoS), it is important to control access to the switching. Toward this goal, we develop high-performance access control and bandwidth allocation for ATM and STM switching, respectively. With respect to access control for ATM switching, we propose an input access scheme for input-queued ATM multicast switches, achieving high system throughput, low packet delay and packet loss probability. Multicast and unicast packets of each input port are separately queued. Multicast queues take priority over the unicast queues, and both types of queues are fairly served based on a cyclic-priority access discipline. In particular, each unicast queue is handled on a window-service basis, and each multicast packet is switched in a one-shot scheduling manner. To evaluate the performance of the access scheme, we propose approximate analyses based on a simplified cyclic-priority model for finite-buffer unichannel and multichannel switches possessing Bernoulli multicast and unicast arrivals, with window-service (for unicasting) and one-shot scheduling (for multicasting) both taken into account. We also show simulation results to demonstrate the accuracy of the approximate analyses and the superiority of the scheme over existing schemes with respect to normalized system throughput, mean packet delay, and packet loss probability. As for bandwidth allocation for STM switching, we focus on the bandwidth assignment to voice and data traffic. Our goal is to analytically determine optimal bandwidth allocated to voice and data traffic by means of a queueing model with heterogeneous arrivals and multiple designated channels. The accuracy of analytical results is confirmed by simulation results. On the basis of the analysis, we propose a polynomial-bounded algorithm to construct the bandwidth assignment paradigms for the assignment of network bandwidth to voice and data traffic in an effort to guarantee minimal data delay and voice call blocking probability. Therefore, the resulting bandwidth assignment assures QoSs in terms of data delay and voice-call blocking probability under various network loads. Consequently, we can combine the input access scheme and the bandwidth assignment paradigms in an integrated and efficient manner to achieve improved network utilization and satisfy diverse QoS requirements.