Introduction to Networking for NonStop S-Series Servers

TCP/IP Network Connections
Introduction to Networking for HP NonStop S-Series Servers—520670-005
10-9
Parallel Library TCP/IP Architecture and Features
to the correct processor from the adapter, Parallel Library TCP/IP eliminates the
message-system hop that occurred between processes in the NonStop TCP/IP
architecture.
By eliminating message system hops, Parallel Library TCP/IP reduces the total
path-length from the application to the wire. This path-length reduction reduces
individual request latency. In addition, more requests per second can be serviced with
the same processor cost, resulting in higher throughput.
Single IP Host
In NonStop TCP/IP, if you were to run multiple process instances of a listening
application in multiple processors (to increase computing power), you needed a
different TCP/IP process, (one per listening application process instance), in each
processor. Each of those TCP/IP processes required a unique physical port (PIF) and
presented a unique IP host to the outside world.
By contrast, in Parallel Library TCP/IP, you can run multiple process instances of a
listening application in multiple processors, all sharing the same PIF. ServerNet allows
all processors in a clustered system to access the same PIF; Parallel Library TCP/IP
allows applications in different processors to access the same PIF and share a
common listening TCP port number. Hence, in Parallel Library TCP/IP, multiple
application process instances running in different processors can be presented to the
outside world as a single IP host.
Ethernet Failover
Ethernet failover, available as of the G06.10 RVU of Parallel Library TCP/IP, provides
fault tolerance at the adapter level. With Ethernet failover, you can configure your
network to continue running if an adapter fails or during maintenance and replacement
of an adapter.
Ethernet failover allows TCP and UDP sessions to continue operating if there are
cabling or adapter failures. With Ethernet failover, network traffic automatically
migrates from the faulty logical interface (LIF) to the working LIF.
Ethernet failover requires two logical interfaces or LIFs. To implement Ethernet
failover, you can configure one Ethernet 4 ServerNet adapter (E4SA) or one Gigabit
Ethernet 4-port ServerNet adapter (G4SAs), which contain four Ethernet ports—each
of which is configured through a LIF—or two Fast Ethernet ServerNet adapters
(FESAs) or two Gigabit Ethernet ServerNet adapters (GESAs). (Two FESAs or GESAs
are needed because these adapters only have one Ethernet port and one LIF.)