Exchange methods between NBMA and BMA

The exchange between NBMA and BMA is a network term translated as "non-broadcast Multi-Channel Access", so BMA is "broadcast multi-channel access. The two concepts describe the characteristics of the data link layer, and one N or less can affect the design of the link layer switch. In general, NBMA is used by the switch to determine how to forward data frames, while BMA is used by the host to determine whether to accept a certain data frame. The switch only plays a secondary optimization role, for example, filter unnecessary broadcasts. The NBMA switch must be manually configured (even if there is a so-called dynamic configuration protocol, the protocol operation policy is also set). On the contrary, the BMA switch, such as the Ethernet switch, is passively learned. Www.2cto.com 1. Although everyone is connected to a link, this does not mean that data frames can be anywhere. In the NBMA network, the control is on the switch, and the terminal host can only reach the target host through the switch. That is to say, only when the "data frame from this port has passed some test" is configured on the switch, the data frame can pass only when the policy can be issued from that port. For the NBMA network, the implementation of the switch is more complex, because it has a lot of built-in protocol layer logic, rather than a passive learning. It may be better if we use several NBMA instances. For NBMA, Frame Relay and ATM are the most typical technologies. The widely used technologies are ATM. We know that ATM is very complicated and much more complicated than Ethernet, I think it is more complicated than IP. An important concept of ATM is virtual circuit and virtual channel. virtual circuit must be established before communication. The work of the ATM switch is to exchange virtual paths so that data frames can be transmitted in the configured virtual channel, A specific data frame can only move forward along one virtual channel and will not arrive at another virtual channel. The virtual channel configuration of the ATM is performed on the switch, that is, the switch controls how data frames are forwarded, and the terminal host can only passively accept this arrangement. The advantage of NBMA is its controllability, unlike BMA. For the WAN, BMA is rarely used, and most of them use NBMA. Because the WAN forwarding policy is quite complex, the link layer is generally forwarded by a specific target under the guidance of BGP, there must be no arbitrary broadcast, which is related to the mandatory policies of various AS. If any broadcast occurs, the WAN will not be secure. The security here may be highly related to politics and national security. 2. in fact, the BMA network is neither an exchange network nor an exchange network. The logical topology of the BMA network is always a bus-type topology. The concept of switching is introduced in the BMA network to optimize the network. Taking Ethernet as an example, the original Ethernet is bus-type. Later, in order to facilitate expansion, the HUB was introduced and a bridge was used to connect heterogeneous Ethernet networks, later, Ethernet switches were widely used in high-performance and low-cost twisted pair wires, and bus-based Coaxial Lines gradually exited. The biggest advantage of an Ethernet switch is its learning function. This learning is a kind of passive learning. Although there is no secret in the BMA network, due to the existence of a Learning switch, secrets are confined to mutually accepted communication parties. Throughout the history, we can find that the Ethernet switch is a latecomer, although it really subverts the Ethernet architecture, so that the later Gigabit Ethernet is only compatible with the traditional 802.3 for the sake of compatibility, what is the essence of Ethernet? This is the ARP Protocol. ARP requests are a broadcast. This broadcast confirms that Ethernet is the original broadcast network. Any communication has to use the arp protocol for address resolution (without considering static arp ing configuration). This address resolution and subsequent unicast communication give the switch a learning opportunity, knowing the port on which a MAC address is accessible to the vswitch, The vswitch can filter out unnecessary broadcasts.

This kind of learning is passive boosting. Even if there is no such learning mechanism, data packets can still reach the target, because the implementation of the BMA network on the end host, rather than the switch, for Ethernet, the default data frame will reach any host on the same link, and the host will filter out the data frame of the target MAC instead of the local host. This is the essence of BMA-that is, it is implemented on the host, and the switch only helps, when a vswitch is unable to judge (such as aging MAC/port ing entries), it will return to the essence of BMA, that is, broadcast data frames on each port. IPv6 removes arp broadcast, which in turn completely destroys the foundation of Ethernet broadcast. However, even so, a data frame can still change the target MAC address to a broadcast address. Before the standard changes, when the vswitch is still in the blind state for multicast, the switch still forwards the data frame on all ports.