ATM網路訊務控制方法之研究

Abstract

寬頻整合服務數位網路(B_ISDN)為了能夠同時支援不同服務品質的需求﹐網路必須擁有足夠的頻寬、有效率的頻寬配置､擴充性佳及能提供各式各樣的服務需求。非同步傳輸模式（ATM）利用精簡的通訊協定﹐擁有高速的資料傳輸速率﹐而其採用Statistical Multiplexing之技術﹐提高了頻寬的使用效率及彈性的配置頻寬。因此﹐B_ISDN之傳輸技術選定非同步傳輸模式以提供多樣化的服務。多樣化服務的分類主要是因應訊務之特性及對網路服務品質的需求﹐B_ISDN所提供多樣化服務包括﹕固定位元速率服務(CBR)､可變位元速率服務(VBR)､可利用位元速率服務(ABR)､不特定位元速率服務(UBR)及保證框架速率服務(GFR)。網路為了提供多樣化服務適當的服務品質﹐有效的訊務控制是必要的﹐而訊務控制的設計會依據不同類別的服務而有所差異。因此我們分別針對固定位元速率服務､可變位元速率服務､可利用位元速率服務､及保證框架速率服務提出有效的訊務控制方法。 對於固定位元速率服務﹐我們提出了一Adaptive Delay-Jitter Control (ADJC)的訊務控制﹐此方法保證固定位元速率服務之延遲效能(Delay Performance)。ADJC可以根據所需的延遲效能作適當的控制。相較於過去文獻所提出的方法﹐ADJC能夠提供較佳的延遲效能而且易於實作。把聲音視為CBR訊務以AAL1來傳送將會造成頻寬的浪費﹐以可變位元速率服務方式來傳送聲音資料是較佳的選擇﹐它使網路資源的配置將可依據實際的需要。因此﹐我們設計一個有有效的可變位元速率訊務控制以滿足聲音服務所需的服務品質。我們提出了兩種方法﹕Circuit Emulation Traffic Control(CETC)及Adaptive Priority Traffic Control(APTC)。CETC將聲音視為固定位元速率服務訊務而且保證其所需的服務品質(Delay- Jitter)。APTC則將聲音視為可變位元速率服務訊務﹐利用動態緩衝區配置方法依據實際的訊務量來配置緩衝區﹐並且利用adaptive priority queuing technique滿足不同訊務之延遲需求。同時APTC亦提高了網路的multiplexing gain。由於ABR訊務具有bursty與unpredictable的特性﹐相較於其它服務﹐其訊務控制更加困難。針對ABR服務﹐我們提出Offset Proportional Rate Control Algorithm(OPRCA)。所提出的方法不僅可達到weighted fair sharing﹐而且有較高的效能及較低的延遲。此外﹐我們亦提出一緩衝區管理方法以避免Head-of-Line(HOL) blocking。OPRCA結合link-by-link feedback提供了端點到端點速率的控制及ABR服務之品質要求。對於保證框架速率服務﹐我們提出了一packet-discard push-out (PDPO)的訊務控制﹐PDPO提供保證框架速率服務品質而且有效的解決封包切割所引發的問題﹐進而提升TCP的傳輸效能。我們亦提出可行的製作方法。

Broadband integrated service digital network(B-ISDN) utilizing asynchronous transfer mode(ATM) is intended to support a wide variety of services and applications, as well as to satisfy various user quality needs. In ATM networks, statistical multiplexing provides high bandwidth utilization and supports a variety of traffic patterns including bursty traffic.The different types of services are categorized into following service classes:constant bit rate(CBR), variable bit rate(VBR), available bit rate(ABR),unspecified bit rate(UBR) and UBR+/GFR(Guaranteed Frame rate). However, the objectives of ATM technique can not be realized without effective traffic control approaches. Generally, functions of traffic control are structured differently for each service category. In this dissertation, effective traffic control approaches for CBR, real-time VBR, ABR and UBR+ services are investigated. For CBR services, we present an adaptive delay-jitter control(ADJC) method that guarantees the delay bounds of CBR services in ATM networks. Through mathematical derivation, we have demonstrated that ADJC features a significant improvement over existed approaches in jitter delay and feasibility of implementation as well as provides a better delay-jitter control. Most voice services are transported through AAL1, in which voice calls are handled as CBR traffic. This causes a lot of wasted bandwidth.Sending voice through VBR may be a better alternative, because it allows the network to allocate voice bandwidth on an as-needed basis. We propose two traffic control approaches, a circuit emulation traffic control(CETC) and an adaptive priority traffic control(APTC) for supporting voice services. In CETC, the service discipline guarantees the quality of service(QOS) for voice circuits. We also present an APTC approach which uses a dynamic buffer allocation scheme to adjust the buffer size based on the real traffic demand, as well as employs an adaptive priority queuing technique to handle various delay requirements for VBR voice traffic. It provides an adequate QOS for voice circuits in addition to improving the multiplexing gain. Due to bursty and unpredictable pattern of an ABR data stream, its traffic control is more challenging than other services. we present an improved ABR traffic control approach, called Offset Proportional Rate Control Algorithm(OPRCA). The proposed approach achieves high link utilization, low delay and weighted fair sharing among contending sources. Moreover, OPRCA combines with the link-by-link feedback scheme which employs a buffering scheme to avoid Head-of-Line(HOL) blocking. To provide GFR service guarantees, we propose a packet-discard push-out(PDPO) control approach which is constituted of the selective packet-discard with available-buffer-space tracking scheme and a packet-based push-out buffering scheme. PDPO fulfills the requirements of GFR service and alleviates the effect of fragmentation as well as improves the TCP throughput using the common FIFO cell scheduling. A feasible realization approach is also addressed.