device(s) being tested. A partial list of the metrics includes: throughput, frame latency, throughput fairness, frame loss ratio, maximum frame burst size, and call establishment. A brief overview of these metrics follows.
Throughput
Three different frame-level throughput metrics are defined.Lossless throughputis the maximum rate at which none of the offered frames is dropped by the system. Thus, the lossless throughput is the highest load at which the count of the output frames equals the count of the input frames. Peak throughputis the maximum rate at which the system operates regardless of frames dropped. Thus, the peak throughput is the maximum throughput that can be achieved in spite of the losses. The maximum rate can actually occur when the loss is not zero. Also, note that the peak throughput may equal the lossless throughput in some cases. Full-load throughput is the rate at which the system operates when the input links are loaded at 100% of their capacity. In all cases, only frames that are received completely without errors are included in frame-level throughput.
A model graph of throughput vs. input rate is shown in Figure 1. Levelxdefines the lossless throughput, levelydefines the peak throughput and levelzdefines the full-load throughput.
Figure 1: Peak,
lossless and full-load throughput
Throughput is expressed in effective bits/sec, counting only bits from AAL payloads excluding the overhead introduced by the ATM technology and transmission systems. This is preferred over specifying it in frames/sec or cells/sec. Frames/sec requires specifying the frame size. The throughput values in frames/sec at various frame sizes cannot be compared without first being converted into bits/sec. Cells/sec is not a good unit for frame-level performance since the cells are not seen by the user.
Before starting measurements, a number of VCCs (or VPCs), called foreground VCs, are established through the system. Foreground VCs are used to transfer only the traffic whose performance is being measured. That traffic is referred as the foreground traffic.
Foreground traffic is specified by the type of foreground VC, connection configuration, service class, arrival patterns, frame length, and input rate.
Foreground VCs can be permanent or switched, virtual path or virtual channel connections, established between ports on the same network module on the switch, or between ports on different network modules, or between ports on different switching fabrics.
A system withn ports is tested for the following connection configurations:
· n-to-n straight: Input from one port exits to another port. This represents almost no path interference among VCs. There aren VCs.
· n-to-(n–1) full cross: Input from each port is divided equally to exit on each of other (n–1) ports. This represents intense competition for the switching fabric by VCs. There aren×(n–1) VCs.
· n-to-m partial cross: Input from each port is divided equally to exit on othermports (1<m<n–1). This represents partial competition for the switching fabric by VCs. There aren×mVCs. Note that n-to-n straight and n-to-(n–1) full cross are special cases of n-to-m partial cross withm=1 andm=n–1, respectively.
· k-to-1: Input fromk(1<k<n) ports is destined to one output port. This stresses the output port logic. There arekVCs.
· 1-to-(n–1): Input from one port is multicast to all other output ports. This tests the multicast performance of the switch. There is only one VC.
Different connection configurations are illustrated in Figure 2, where each configuration includes one ATM switch with four ports, with their input components shown on the left and their output components shown on the right.
The following service classes, arrival patterns and frame lengths for foreground traffic are used for testing:
· UBR service class: Traffic consists of equally spaced frames of fixed length. Measurements are performed at AAL payload size of 64 B, 1518 B, 9188 B and 64 kB.
· Variable length frames and other arrival patterns (e.g. self-similar) are studied
· ABR service class is under study.
Higher priority traffic like VBR or CBR can act as background traffic. Details of background traffic characteristics have not yet been defined.
The input rate of foreground traffic is expressed in the effective bits/sec, counting only bits from AAL payloads excluding the overhead introduced by the ATM technology and transmission systems.
It is obvious that testing larger systems, e.g., switches with larger number of ports, could require very extensive (and expensive) measurement equipment. Hence, we introduce scalable test configurations for throughput measurements that require only one ATM monitor with one generator/analyzer pair. Figure 3 presents a sample test configuration for an ATM switch with 8 ports in an 8-to-2 partial cross connection configuration. The configuration emulates 16 foreground VCs.
There is one link between the ATM monitor and the switch. The other seven ports have external loopbacks. A loopback on the given port causes the frames transmitted over the output of the port to be received by the input of the same port.
The test configuration in Figure 3 assumes two network modules in the switch, with switch ports P0-P3 in one network module and switch ports P4-P7 in the another network module. In this case, foreground VCs are always established from a port in one network module to a port in another network module.
Figure 2: Connection configurations for throughput measurements