Visara LINCS Features - Manual (Page 16)

6
LINCS Features
707021-001
there, the NN sends the request to all of its adjacent network nodes. The send process is
repeated by the adjacent nodes, until the LU is found. At that time, APPN caches the location
of the found LU, so it will not have to go through the search phase if that LU is requested
again. If more than one possible route is found, APPN selects the best path of those available
which meet the requested requirements.
Network Topology
APPN allows any network topology. Each NN can be directly connected to every other NN,
connected through a single routing hub, connected in a hierarchical network design, or any
combination of these. LINCS APPN maintains information about all NNs and intermediate
routing Transmission Groups (TG) within the network in a “Safe Stored” Topology Database.
LINCS exchanges network information with other network nodes to maintain up-to-date
topology information, which includes data about NNs, plus their connections to Virtual Routing
Nodes and other NNs. Unlike TCP/IP, topology information is exchanged only when a topology
change occurs, which reduces network management traffic significantly. Also, the topology
data base contains information about network nodes only (Information about LEN and ENs is
obtained from APPN’s directory service) which reduces the size of the database.
Safe Store
All LINCS nodes are equipped with hard disks, so LINCS can save network information,
which is known as the APPN Safe Store feature. Safe Stores occur only when updates to the
directory or topology database have occurred. LINCS checks whether updates have occurred
on heuristically determined 1 to 10 minute intervals. If LINCS APPN is stopped and restarted,
only topology and directory updates that have occurred since the last Safe Store need to be
obtained from the adjacent end node. This greatly reduces the amount of network management
data traffic.
Intermediate Session Routing (ISR)
LINCS APPN supports networks using Intermediate Session Routing. ISR provides connection-
oriented routing for LEN and EN sessions and for connection to adjacent NNs. Additionally
ISR provides adaptive pacing and segmentation and reassembly of APPN network message
units when required by partners. ISR routes session traffic at the Path Information Unit level.
High Performance Routing (HPR)
High Performance Routing (HPR) is an extension to APPN, which uses the same link types
that base APPN supports. HPR adds reliability to the network, so link-level error recovery is
only recommended on links with high error rates. HPR requires that LLC does not support
link-level error recovery. HPR requires that X.25, Channel/SNA, SDLC, SDLC/DAP, do operate
with link-level error recovery.
HPR provides dynamic rerouting for APPN, so sessions can survive link failures, while
maintaining the deterministic stability and class-of-service associated with ISR. HPR can
coexist and interoperate with all existing levels of APPN nodes. HPR nodes can distinguish
between base APPN and HPR nodes and links. APPN nodes view HPR nodes and links as
base APPN nodes and links. HPR uses the existing APPN route selection algorithm for
route selection.