Which two options are design considerations when introducing FCoE into an existing network? (Choose two)

Question

Options

• AThe FCoE QoS markings may overlap with call signaling QoS markings
• BOptical cabling is needed to transmit FCoE traffic between a server and its directly connected
• CThe existing network must support a MTU of 3280 bytes
• DTwinaxial cabling can be used to transmit FCoE traffic between a server and its directly connected
• EAll the servers in the data center must be retrofitted with converged Network Adapters

Explanation

A and D are the correct answers.

FCoE requires lossless Ethernet, which means specific QoS priority classes (typically CoS 3) must be reserved for storage traffic using Priority Flow Control (PFC). This can collide with existing voice/unified communications deployments that also use CoS 3 for call signaling - making QoS overlap (A) a genuine design concern that must be planned for carefully. Twinaxial (DAC) cabling (D) is a valid and common physical layer option for FCoE between servers and top-of-rack switches at short distances (up to ~10 meters), so it is a real consideration when planning cabling.

Why the others are wrong:

• B is wrong because optical fiber is not required - Twinax copper is explicitly supported, which is exactly what D confirms.
• C is wrong because FCoE requires a minimum MTU of 2158 bytes (sometimes rounded to 2500); 3280 is not a standard FCoE MTU figure.
• E is wrong because only servers that need FCoE storage access require a Converged Network Adapter (CNA) - you do not need to retrofit every server in the data center.

Memory tip: Think "QoS clash and cable choice" - the two design headaches when dropping FCoE into an existing network are ensuring your QoS markings don't fight with voice traffic (A), and knowing that cheap Twinax copper is a valid cable option, not just fiber (D).

No community discussion yet for this question.