ATM access-shaping and bandwidth-tuning of DQDB egress traffic
暂无分享,去创建一个
This paper is concerned with the asynchronous transfer mode (ATM) network resources required by distributed queue dual bus (DQDB) traffic. The set up used for performing measurements under different traffic load scenarios involves a DQDB network and a set of ATM traffic control and enforcement modules such as a virtual shaper (VS) and a usage parameter control (UPC) device. To reassure the measurements tractability, we have chosen the set of traffic descriptors currently promoted for standardization in the international fora. These include typical parameters like the peak cell rate (PCR) of the source, but also a number of novel ones like the sustainable cell rate (SCR) that can represent the connection scale bandwidth use since its definition takes into account statistical characteristics of the source. The paper studies the characteristics of the traffic appearing in the VS, i.e., the effect of the DQDB multiplexing discipline to native traffic streams is shown. Doing so, we are able to determine the range of the traffic descriptors values (PCR, SCR, MBS) towards acceptable cell loss and delay, but also to recommend on the operation and performance of the VS in terms of cell delay variation (CDV) and buffer space.
[1] Murad S. Taqqu,et al. On the Self-Similar Nature of Ethernet Traffic , 1993, SIGCOMM.
[2] James Roberts. Virtual spacing for flexible traffic control , 1994 .
[3] John D. Angelopoulos,et al. Service Integration in DQDB MANs and Impact on ATM Interworking Performance , 1994 .
[4] Corporate. IEEE Std 802.6-1990, IEEE Standards for Local and Metropolitan Area Networks: Distributed Queue Dual Bus (Dqdb) Subnetwork of a Metropolitan Area Network (Man) , 1991 .