Multiple Shared Ethernet Adapter Configuration - IBM p5 590 System Handbook

When the shared Ethernet receives IP (or IPv6) packets that are larger than the
MTU of the adapter that the packet is forwarded through, either IP fragmentation
is performed and the fragments forwarded or an ICMP packet too big message is
returned to the source when the packet cannot be fragmented.
Theoretically, one adapter can act as the only contact with external networks for
all client partitions. Depending on the number of client partitions and the network
load they produce performance can become a critical issue. Because the Shared
Ethernet Adapter is dependant on virtual I/O, it consumes processor time for all
communications. A significant amount of CPU load can be generated by the use
of virtual Ethernet and Shared Ethernet Adapter.
There are several different ways to configure physical and virtual Ethernet
adapters into Shared Ethernet Adapters to maximize throughput.
Using Link Aggregation (EtherChannel), several physical network adapter
can be aggregated. See 3.4.2, "Using Link Aggregation (EtherChannel) to
external networks" on page 77 for more details.
Using several Shared Ethernet Adapters provides more queues and more
performance. An example for this configuration is shown in Figure 3-9.
Other aspects which have to be taken into consideration are availability and the
possibility to connect to different networks.
Figure 3-9 Multiple Shared Ethernet Adapter configuration
IBM Eserver p5 590 and 595 System Handbook
76