Ethernet used to be found only in corporate LANs but is now proliferating into the carrier network, from access and metro area network to the backbone. It has some pesky problems you need to know about: For instance, what is duplex, and what is the difference between half and full duplex? What is a duplex mismatch? How can auto-negotiation help or hurt your network? How can you resolve duplex conflicts? Here's some help.
Duplex mismatch problems -- caused by two ends of the Ethernet attempting a full-duplex connection, resulting in packet loss -- simply will not go away. After a decade of plaguing IP networks, duplex conflicts still seem to be the single worst source of performance degradation. Trivial to fix but ridiculously hard to identify and localise, mismatches recur frequently, as interfaces go up and down over time and network hosts are updated and changed. Find out why duplex mismatches are difficult to conquer using auto-negotiation.
part 2: Stop the conflicts
Today, the majority of modern cablling supports full-duplex, and there is a significantly diminished need for half-duplex -- or at least, that is what one would like to think. This article examines extenuating circumstances where half-duplex may be appropriate and offers best practices where its use may still be needed.
This article helps networking engineers understand the issues behind duplex mismatch, one of the most troublesome problems in today's Ethernet networks. Learn why duplex mismatch occurs and how you can prevent it with autonegotiation and hard coding.
Learn how to detect duplex mismatch in the first place. Once you identify the problem, you can turn your attention to solving this annoying performance issues on the Ethernet network.