![]() The loss of a single fragment results in all the fragments having to be resent where a reliable transport layer protocol such as TCP is in use (in fact the sender resends one packet and fragmentation occurs once again).IP And higher layer protocols to work across variable and diverse network paths and mediums without the need and overhead of a path discovery protocol (but see the PMTUD section).įragmentation has a number of drawbacks which result in it’s use being avoided where possible, primarily:.Transport layer protocols to be ignorant of the underlying network architecture, reducing overheads.As I’ve made clear, its the data within it that’s fragmented – the original packet is actually discarded. Typically the language used when discussing fragmentation implies the original packet itself is fragmented. This is called Fragmentation and the data in these fragments is then typically reassembled when they reach their destination. This means it is broken into pieces carried within new packets (fragments) that are equal to or smaller than the lower MTU. In these cases, if the packet size exceeds the lower MTU the data in the packet must be fragmented (if possible). ![]() ![]() However, packets that are initially transmitted over a network supporting one MTU may need be routed across networks (such as a WAN or VPN tunnel) with a smaller MTU. A typical contemporary office, campus or data centre network provided over Ethernet will have 1500 byte MTUs. This size is determined by the network’s data link and IP Maximum Transmission Units (MTUs) which are usually the same. When a host sends an IP packet onto the network it cannot be larger than the maximum size supported by that local network.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |