Black Box Modular Switches, LE14XXA manual Frame Buffering and Latency

ack Box Modular SwitchesInstallation and User Guide (8/99)

The RJ-45 ports run at 10/100Mbps with N-wayauto-negotiation capability, whereas e fiber port runs at 100Mbps with half- or full-duplex capability manually selected. The fault condition is full-duplex. Internal jumpers settings allow technicians to set the 100Mb ber port to half-duplex mode. (See Section 3.4).

n LE14XXA Combo 4-port modules, there are three LED’s for each RJ-45 port, which indicate atus as described for the LE1426C in Section 2.2.4 above. The fiber port will run at 100Mbs eed at all times, and has LEDs that indicate status the same way as described for the Fiber 4- rt modules in Section 2.2.2 above.

2.6Frame Buffering and Latency

The Black Box LE14XXA-Series are store-and-forward switches. Each frame (or cket) is loaded into the Switch’s memory and inspected before forwarding can occur. This chnique ensures that all forwarded frames are of a valid length and have the correct CRC, i.e., e good packets. This eliminates the propagation of bad packets, enabling all of the available ndwidth to be used for valid information.

While other switching technologies such as "cut-through" or "express" impose minimal ame latency, they will also permit bad frames to propagate out to the Ethernet segments nnected. The "cut-through" technique permits collision fragment frames, which are a result of te collisions, to be forwarded to add to the network traffic. Since there is no way to filter ames with a bad CRC (the entire frame must be present in order for CRC to be calculated), the sult of indiscriminate cut-through forwarding is greater traffic congestion, especially at peak tivity. Since collisions and bad packets are more likely when traffic is heavy, the result of ore-and-forward operation is that more bandwidth is available for good packets when the traffic ad is greatest.

To minimize the possibility of dropping frames on congested ports, each Black-Box E14XXA-Series Switches dynamically allocates buffer space from an 8 MB memory pool, suring that heavily used ports receive very large buffer space for packet storage. (Many other witches have their packet buffer storage space divided evenly across all ports, resulting in a mall, fixed number of packets to be stored per port. When the port buffer fills up, dropped ckets result .) The other two LE14XXA-Series Switches LE1408A and LE1416A dynamically locates buffer from an 4MB memory pool. This dynamic buffer allocation provides the pability for the maximum resources of the LE14XXA-Series unit to be applied to all traffic ads, even when the traffic activity is unbalanced across the ports. Since the traffic on an erating network is constantly varying in packet density per port and in aggregate density, the

10