GLS provides a network configuration based on multiple NICs connecting to the same segment.
When designing a network configuration, it is important to take into consideration any quantitative restrictions, whether or not NICs are shared, and whether or not VLAN tagging is used.
Standard Configurations
The network configurations usually used with GLS are as follows:
Single Virtual Adapter without Shared NICs
Multiple Virtual Adapters without Shared NICs
Shared NICs (All NICs Shared by Virtual Adapters Using VLAN Tagging)
Shared NICs (Some NICs Shared by Virtual Adapters Using VLAN Tagging)
The main features of these configurations are listed below.
Single Virtual Adapter without Shared NICs
This is the configuration to use when a single network is duplicated.
Multiple Virtual Adapters without Shared NICs
This is the configuration to use when each user application resides on a different LAN. This configuration is also suitable for three-tier application servers and for separating LAN traffic for user operations and administrative operations.
Shared NICs (All NICs Shared by Virtual Adapters Using VLAN Tagging)
This is the configuration to use in environments such as a blade server system, where there is a limit on the number of NICs that can be used and the environment spans multiple networks.
Shared NICs (Some NICs Shared by Virtual Adapters Using VLAN Tagging)
This is the configuration to use when combining any of the three configurations described above.
Note
You cannot create partial NIC shares with multiple virtual adapters in GLS.
In order to share NICs, you must set different VLAN tags for each virtual adapter and each NIC must connect to a different VLAN.
For more details on the above patterns, refer to "Appendix A Sample GLS Configurations."
