Brocade Fabric OS Administrators Guide (Supporting Fabric OS v7.3.0) - Manual (Page 391)

Supported configurations for bottleneck detection
The following configuration rules apply to bottleneck detection:
• Bottleneck detection is supported only on the following Fibre Channel port types:
‐ E_Ports
‐ EX_Ports
‐ F_Ports
‐ FL_Ports
‐ F_Port and E_Port trunks
‐ Long distance E_Ports
• Bottleneck detection is not supported on either SIM ports or FCoE ports.
• Bottleneck detection is supported on 4 Gbps, 8 Gbps, and 16 Gbps platforms, including those
running at 10 Gbps speeds.
• Bottleneck detection is supported in Access Gateway mode.
• Bottleneck detection is supported whether Virtual Fabrics is enabled or disabled. In VF mode,
bottleneck detection is supported on all fabrics, including the base fabric. Refer to Virtual Fabrics
considerations for bottleneck detection on page 392 for additional information on using bottleneck
detection in VF mode.
Limitations of bottleneck detection
The bottleneck detection feature detects latency bottlenecks only at the point of egress, not ingress. For
example, for an E_Port only the traffic egressing the port is monitored. For FCoE ports at the FC to
DCB (Data Center Bridging) boundary, traffic going from FC to DCB is monitored, but traffic going from
DCB to FC is not monitored.
ATTENTION
Latency bottleneck detection is not recommended for link utilizations above 85 percent.
High availability considerations for bottleneck detection
The bottleneck detection configuration is maintained across a failover or reboot; however, bottleneck
statistics collected are lost.
Upgrade and downgrade considerations for bottleneck detection
The bottleneck detection configuration is persistent across firmware upgrades and downgrades.
The sub-second latency criterion parameter settings are not preserved on a downgrade to firmware
versions earlier than Fabric OS 7.0.0. If you downgrade to a firmware version earlier than Fabric OS
7.0.0 and then upgrade to a more recent version of Fabric OS, these settings revert to their default
values.
Trunking considerations for bottleneck detection
A trunk behaves like a single port. Both latency and congestion bottlenecks are reported on the master
port only, but apply to the entire trunk.
Supported configurations for bottleneck detection
Fabric OS Administrators Guide 391
53-1003130-01