Distributed Name Service (DNS) Management Operations Manual

The Need for Better Name Management
Introducing Distributed Name Service
1–2 31258 Tandem Computers Incorporated
Briefly, DNS provides facilities that:
Maintain names of objects controlled by Tandem systems and other types of
systems
Provide alternative names (aliases) for objects
Translate an alias to a subsystem-object names, allowing command interpreters to
accept meaningful names
Translate subsystem-object names to aliases, allowing event processing
applications to report meaningful names to operators
Organize objects into groups
Provide a single name for a set of objects
Translate a group name to the names of the members of that group, allowing
network management applications (NMAs) to implement group-oriented
commands
The Need for Better
Name Management
As networks of Tandem computer systems become larger and more complex, the need
for a better naming system becomes important. Names for objects should meet two
general requirements:
They should be easy to read, understand, and remember. Most users like to select
names that describe in some way the thing to which they refer. For instance,
NEWHIRE can be the name of a program that processes records concerning a
company’s new employees, and INFILE can be the name for an input file used by
an application program.
They should be independent of locations or addresses in the network. This
property lets you move an object to a new location in a distributed system without
changing its name and without changing the existing references to the object.
Typically, names within a system or a network only partly meet the above
requirements. Names defined by subsystems, such as Pathway names, are generally
easy for people to read and remember, but they are valid only within the context of
applications using Pathway.
To refer to a terminal named TERMX, you must not only know that TERMX is a
Pathway object name for a terminal, but you must also know to which Pathway
monitor process TERMX is defined.