ccna r&s book

ccna r&s book.1. Store-and-forwardStore-and-forward switching is the most basic switching method. When using store-and-forward switching, the Layer 2 device must add the entire frame to the buffer of the inbound interface and check the FCS (checksum) of the frame before performing any other processing. To check the FCS (often referred to as cyclic redundancy checksum [CRC]), the Layer 2 device calculates the CRC as if it were the source device and compares it to the CRC value in the frame. If they are the same, the Layer 2 device considers the frame to be valid and can begin processing it, which involves forwarding the frame out of the port to the destination. If the FCS value in the frame is different from the calculated value, the Layer 2 device will discard the frame.Bridges support only the store-and-forward switching method. All switches support store-and-forward, but some switches (such as older Cisco Catalyst 1900 series switches) also support other switching methods, depending on the switch model and manufacturer.2. Pass-ThroughSome switches, such as the older Catalyst 1900 series switches (now retired and no longer sold by Cisco), support pass-through switching. When using pass-through switching, the switch makes a switching decision after reading the first part of the frame. After the switch reads the destination MAC address (8-byte leading code and 6-byte MAC address), it starts forwarding the frame (although the rest of the frame may be coming in through the interface at this point). One of the advantages of straight-through switching is that it is faster, but the biggest problem is that the switch may exchange invalid frames, because the frame header may be fine, but the rest of the contents may be damaged by delay conflicts.Most vendors address this problem by supporting a dynamic switching method. When performing straight-through switching, the switch still checks the CRC of the frames as they are exchanged, in order to find invalid frames. Even if the frame is invalid, it is exchanged, but the switch records the number of invalid frames. If the number of invalid frames reaches a specified threshold within a given time, the switch dynamically switches from straight-through switching to store-and-forward switching. However, whether this feature is provided in a switching product is entirely vendor dependent.3. Fragmentation-freeFragmentation-free switching is a modified version of pass-through switching. While straight-through switching makes the forwarding decision after reading the frame's destination MAC address field, fragment-free switching determines that the frame has at least 64 bytes before switching it (valid Ethernet frames have a minimum of 64 bytes) Fragment-free switching is designed to reduce the number of switched Ethernetwork stub frames (frames shorter than 64 bytes). Fragment-free switching is sometimes referred to as modified straight-through switching or fragment-free switching.Even with fragment-free switching, the switch may still exchange damaged frames (frames with incorrect FCS) because the switch only checks the first 64 bytes and the FCS is at the end of the frame. To solve this problem, many vendors have implemented the dynamic switching method discussed in the previous section. Most conflicts typically generate fragmented frames, and the use of fragment-free switching methods avoids forwarding these frames, unlike pass-through switching plants