How to Set Ethernet Full and Half DuplexFull Duplex Ethernet
Full duplex provides the means of transmitting and receiving simultaneously on a single wire. Full duplex Ethernet is used between two endpoint like for between switches and servers, between switches and routers etc. Full duplex also permit bandwidth on Ethernet and Fast Ethernet networks very easy and cost effective and also fast.. By using features such as Fast EtherChannel, "bundles" of Fast Ethernet connections can be grouped together to increase bandwidth to increase bandwidth up to 4000%.
Full duplex broadcasting is used in Gigabit Ethernet to increase collective bandwidth ranging from 2 Gbps to 2 Gbps for point to point links and also used to increase distances of for specific type of media. Further more Gigabits EthernetChannel allows creating 8 Gbps connections between switches. Full duplex Ethernet minimize chances of collisions so that CSMS/CD does not need to be used for flow control. . However, a full duplex flow control method has been put forward in the standards committee with flow control as on option clause. This standard is referred to as IEEE 802.3x; it formalizes full duplex technology and it's expected to be supported in future Gigabit Ethernet products. Because of the volume of full duplex 100 mbps network interfaces cards (NICs), it is likely that this standard will realistically apply to Fast Ethernet.Half Duplex 10g Ethernet
For half duplex transmission, CSMA/CD will be utilized to ensure that stations can communicate over a single wire and that collision recovery can take place. CSMS/CD implementation should be same as for Ethernet and Fast Ethernet and also it allows creation of shared Gigabit Ethernet using hub of half duplex via point to point connections.
Because the CSMA/CD protocol is delay sensitive, a bit budget per collision domains must be created. When CSMS/CD is used delay sensitivity is always there however full duplex operation has no delay sensitivity and collision domains are defined as time passes for maximum length frame transmission. This transmission, in turn, governs the maximum separation between two end stations on a shared segment. When network speed grows, frame transmission speed decreases and collision domains diameters is maximum. Budget of collision domain is used as maximum signal delay time for different networking components like repeaters, MAC layer and medium also.
Increase in speed of Ethernet speed depends upon some terms that how you are implementing CSMS/CD. Speed which is more than 100 Mbps packet sizes are shorter than length of slot time in bits. (Slot-time is defined as the unit of the time for Ethernet MAC to handle collisions). Handling slot time problem you may find carrier extensions in Ethernet specifications. Carrier extensions are used to add bits to frames until frames meets the minimum requirements which slot time is needed. Following this method, smaller packet sizes can cover with minimum slot time required and it permit seamless operation with current Ethernet CSMS/CD.
More change to the Ethernet requirement is the adding up of frame bursting. Frame bursting is an optional feature in which, in a CSMA/CD environment, and an end station can transmit a burst of frames over the wire without having to relinquish control. Other devices are put back to stop transmission when there is no idle time on wire. .The transmitting station that is bursting onto the wire fills the inter-frame interval with extension bits such that the wire never appears free to any other end station.