TPDC Manual (G06.24+)

TPDC Overview
Tandem Performance Data Collector (TPDC) Manual—526357-001
1-5
How TPDC Collects Data
Generating the UWS file is processor intensive. To minimize TPDC’s impact on
production system performance, use TPDP (for deferred processing) to have
TPDC generate reports itself. To run either of these components whenever you
want after TPDC collects its data, specify the DEFER or INTERVAL options when
you run TPDC.
How TPDC Collects Data
To serve as a useful tool throughout the performance data collection cycle, TPDC:
1. Determines the hardware and software configurations of the system to be
measured
2. Organizes measurements based on the system hardware and software
configuration
3. Starts the measurements
4. Calculates an efficient sampling interval for devices and processors
5. Samples and records information on NonStop Kernel entities
6. Retrieves data from measurements taken by the Measure product
7. Integrates and normalizes Measure data, and relates it to data from other sources
8. Creates a single consolidated data (UWS) file for subsequent analysis and
reporting
9. Creates structured format entities from the consolidated data file to be used by
other products for further analysis and reporting
Entities, Attributes, and Linkages
Entities are collections of data items about types of resources that exist on NonStop
Kernel systems. Each entity has basic attributes, or data elements, that characterize
the entity. Each attribute, or data element, contains the actual data or data value that
TPDC uses in its relational data model.
A linkage is a relationship or mapping between two sets of data entities. Linkages can
be one-to-one, one-to-many, or many-to-many. In TPDC, linkages are represented as
attributes with -LINK as their suffix. The -LINK attributes are assigned the unique 32-bit
identifier value of the related entity. Valid linkages are greater than zero. If the -LINK
attribute has a zero value, this entity is not linked.