A node that connects two different sections of a HIPPI/GSN network. The two sections can be two different fabrics (such as GSN and HIPPI-800) or two similar sections (such as two groups of GSN endpoints). The node complies with the following:

The receiving element located at the other end of a physical link. An upper-layer entity (host, network–layer interface, or program) that uses the GSN subsystem is sometimes loosely referred to as the destination; however, it is more correct to call these software entities upper-layer protocols (that is, destination ULPs). A “final destination” refers to the element that is the ultimate receiver for a micropacket; an element that receives then retransmits the micropacket (for example, a switch) is simply a destination.

Any component of a HIPPI-6400 fabric or system that is able to receive, process, and send HIPPI-6400 Admin micropackets in a manner that conforms with the HIPPI-6400 standard. Each HIPPI-6400 port on an element contains both a source and a destination. For example, the SuMAC chip in an SGI GSN product is a single-port HIPPI-6400 element.

A final destination or an originating source of GSN traffic. An endpoint may have only one GSN port. A single system may have many endpoints (for example, an Origin module with two SGI GSN products has two endpoints).

All the HIPPI nodes (switches, endpoint devices, extenders) that are physically interconnected and speak the same physical–layer protocol. For example, a GSN fabric is a separate entity from a HIPPI–800 fabric.

One GSN fabric can be logically divided into multiple upper-layer address spaces (that is, networks or LISs). For example, a single GSN fabric can support multiple IP networks. Conversely, one logical network can include members from multiple HIPPI fabrics. For example, an IP network can include members from a GSN (HIPPI-6400) fabric as well as members from a HIPPI-800 fabric.

The string of characters and numerals used to identify a specific item of hardware. This string is assigned by the operating system, and is not configurable. The hardware names for the components/devices in a system can be displayed with the hinv command. For the SGI GSN hardware, this name has the format gsn0, gsn1, gsn2, and so on. See “Assignment of Unit Numbers to Hardware” in Chapter 1 for details on how the numbers are assigned to each hardware device (port).

A logical IP subnet (LIS) service that provides address resolution for IP hosts operating over a HIPPI fabric. The service conforms with the proposed HARP RFC. When the HIPPI fabric supports broadcasting, every IP host provides the service by responding to address resolution requests that are addressed to it. If the GSN fabric does not support broadcast, certain hosts provide the service by maintaining HARP tables and answering address requests from the other hosts (clients) of the LIS.

A number used in HIPPI–6400 Admin micropackets to indicate through how many elements the micropacket should be forwarded. Each time a micropacket exits an element, the hop count is decremented by one. See “GSN Admin Micropackets” in Chapter 1 for further details.

The string of characters and/or numerals used to identify a specific instance of an operating system (that is, a UNIX host). This string is completely customer-configurable; it is created by editing the /etc/sys_id file.

Internet address, and frequently referred to as IP address or IP/ST address. A globally–unique 32-bit number used to identify a network–layer entity that speaks one of the DARPA Internet protocols (for example, TCP, UDP, ICMP) or another supported protocol, such as Scheduled Transfer (ST). This address is completely customer-configurable, but assigned in blocks by a globally-recognized address authority, such as the National Science Foundation's InterNIC. In IRIX, this address is mapped to a network interface name by editing the /etc/config/netif.options file. See the network connection name glossary entry for a user-friendly name for this address.

The ST endpoint that sends the first operation within an ST sequence. The endpoint that acts as initiator during one sequence (for example, the connection setup) can act as the responder in a subsequent sequence (for example, the data movement).

See logical IP subnet.

Also known as LIS. A globally-known (public) logical address space that is defined by an INET network address and a subnet mask. The basic methodology of an LIS is similar to subnet, at the local (site) level; however, the consequences for routing are quite different for the two methodologies. Refer to “Classless Inter-domain Routing” RFC 1519. Also see subnet and subnet mask.

See subnet mask.

Sometimes referred to as alias for IP address or (incorrectly) as hostname. The string of characters and/or numerals used as a user-friendly method for identifying a specific network–layer entity; the string is an alternate (alias) for an INET address. This string is completely customer-configurable; it is created and mapped to an INET address in the /etc/hosts file; it can be used in the /etc/config/netif.options file to configure a network interface. By convention, the network connection name includes the system's hostname. For example, a system with the hostname granite might have network connection names of granite-ef0.rocks.com and granite-gsn0.rocks.com.

The string of characters and numerals used to identify a specific logical network–layer interface (sometimes known as an if_net instantiation). This string is assigned by the operating system and tied to a specific hardware subsystem; it is not customer configurable. For the IRIS GSN product, this name has the format gsn0, gsn1, gsn2, and so on. The network interface name is mapped to a configurable network–layer (for example, INET) address by the /etc/config/netif.options file. See “Assignment of Network Interface to Hardware Device” in Chapter 1 for details on how the numbers are assigned to each GSN network interface.

The ST protocol data unit. It is composed of a 40–byte header and variable–length data ranging from 0 bits to 4 gigabits (illustrated in Figure 1-13). Each ST operation is transmitted as one GSN Message, as illustrated in Figure 1-13.

One section of HIPPI–6400–PH transmission medium (copper or fiber-optic cable) that connects two HIPPI–6400-PH elements. Each element can be either a switch or an endpoint. Each physical link is a full-duplex link composed of two simplex links; each simplex link carries data in only one direction. The two streams of data in the full-duplex link flow in opposite directions. The path (virtual connection) between an original point of transmission (the originating source) and a final point for reception (the final destination) can involve numerous physical links.

The other (not the initiator) ST endpoint participating in an ST sequence.

In the context of the ST protocol, a series of operations that occur in a specific order and accomplish an ST protocol task.

Memory at an ST destination that is reserved for holding one incoming ST Header.

The transmitting element located at one end of a physical link. An upper-layer entity (host, network–layer interface, or program) that uses the GSN subsystem is sometimes loosely referred to as the source; however, it is more correct to call these software entities upper-layer protocols (that is, source ULPs). An “originating source” refers to the element that first transmitted a micropacket; an element that is retransmitting the micropacket (for example, a switch) is simply a source.

Also called subnetwork. A site-defined (private) address space that is carved from a globally-assigned (public) INET network address space. A subnet mask is used to divide the INET network address into many subnets. See subnet mask.

Also called netmask. A customer-configurable value for increasing the number of bits within an INET address that are used for network (netid) identification. The netid is increased by taking 1 or more bits from the host portion (hostid) of the INET address and using them as additional network bits. For example, a basic INET class C address consists of 24 bits of netid and 8 bits of hostid. A site can use a subnet mask to alter their addresses so that 28 bits are used to identify the network and only 4 bits are used to identify hosts.

Use of a subnet mask allows one INET network address to be subdivided into multiple networks (called subnets). A basic class C address allows a site to have up to 254 hosts on one network. If a site increases its netid to 28 bits, it can have up to 15 hosts on each of 16 different networks (subnets).

Within a subnet mask, the bits set to 1 specify the portion of the address used to identify networks, while the bits set to 0 identify the hosts. For example, netmask 0xFFFFFF80 (hexadecimal) provides 25 bits of netid and 7 bits of hostid, while 0xFFFFFFF0 provides 28 bits of netid and 4 bits of hostid. The default netmask for IRIX is 0xFFFFFF00. This default can be changed for each network interface by editing the appropriate /etc/config/ifconfig-#.options file.

A node that is located along the route between two endpoints. GSN traffic passes through the switch on its way to a destination endpoint. A switch must have at least two, and usually has numerous, GSN ports.

The numeral portion of the hardware device name. See “Assignment of Unit Numbers to Hardware” in Chapter 1 for a description of how these are assigned.

Also known as a MAC (media access control) or ethernet address. A globally–unique 48-bit IEEE 802-style number used to identify a hardware device. The highest bits (that is, 47 to 24: the Organizationally Unique Identifier) are assigned by the globally-recognized address authority, IEEE. The lower 24 bits are assigned by the local organization that has been given (assigned) the OUI.