Send documentation comments to mdsfeedback-doc@cisco.com
22-10
Cisco MDS 9000 Family Fabric Manager Configuration Guide
OL-17256-03, Cisco MDS NX-OS Release 4.x
Chapter 22 Configuring Generation 2 and Generation 3 Switching Modules
Buffer Credit Allocation
Ports configured in dedicated rate mode are allocated the required bandwidth to sustain a line rate of
traffic at the maximum configured operating speed, and ports configured in shared mode share the
available remaining bandwidth within the port group. Bandwidth allocation among the shared mode
ports is based on the operational speed of the ports. For example, if four ports operating at speeds 1 Gbps,
1 Gbps, 2 Gbps, and 4 Gbps share bandwidth of 8 Gbps, the ratio of allocation would be 1:1:2:4.
Out-of-Service Interfaces
On supported modules and fabric switches, you might need to allocate all the shared resources for one
or more interfaces to another interface in the port group or module. You can take interfaces out of service
to release shared resources that are needed for dedicated bandwidth. When an interface is taken out of
service, all shared resources are released and made available to the other interface in the port group or
module. These shared resources include bandwidth, rate mode, BB_credits, and extended BB_credits.
All shared resource configurations are returned to their default values when the interface is brought back
into service. Corresponding resources must be made available in order for the port to be successfully
returned to service.
Caution If you need to bring an interface back into service, you might disrupt traffic if you need to release shared
resources from other interfaces in the same port group.
Buffer Credit Allocation
This section describes how buffer credits are allocated to switches and modules, and includes the
following topics:
• Buffer Pools, page 22-10
• BB_Credit Buffers for Switching Modules, page 22-13
• BB_Credit Buffers for Fabric Switches, page 22-22
• Extended BB_Credits, page 22-23
Buffer Pools
In the architecture of Generation 2 and Generation 3 modules, receive buffers shared by a set of ports
are called buffer groups. The receive buffer groups are organized into global and local buffer pools.
The receive buffers allocated from the global buffer pool to be shared by a port group are called a global
recieve buffer pool. Global receive buffer pools include the following buffer groups:
• Reserved internal buffers
• Allocated BB_credit buffers for each Fibre Channel interface (user configured or assigned by
default)
• Common unallocated buffer pool for BB_credits, if any, to be used for additional BB_credits as
needed
• Performance buffers (only used on 12-port 4-Gbps and 4-port 10-Gbps switching modules)
To Next Page
Loading...
Need help?
General
