This document is intended to cover the steps required for customers taking
part in the Size By Example feedback program for deployed production customer
portals running on OracleAS Portal 9.0.2.3 and above.
A size-by-example (SBE) approach requires a set of known samples that may be
used as data-points along the thermometer of system size. The more examples
available for SBE the more accurate the intended implementation will be.
Oracle has the ability to deliver targeted SBE sizing solutions for our
prospective portal customers through reference implementation documents that
outline both our internal deployments and customer's external deployments.
By participating in the SBE program you will be assisting Oracle and other
customers through adding to the Portal Implementation Case Study repository.
These documents are records of real world portal installations implemented by both customers and Oracle.
Using these documents Oracle can be assured
when recommending similar configurations that the proposals have been implemented before and will
provide the performance and functionality unique to the proposed implementation.
Ensure that you follow the advisories and checklist below before
transmitting the packaged response back to Oracle.
Further Guidelines
The process of producing a size-by-example document requires two essential
elements; performance metrics and an architectural overview.
The performance metrics provided by the Portal Performance Scripts cover off
the data required for monitoring the performance of the Oracle software, further
metrics are required for provision of operating system level information.
In terms of Unix operating systems this is simply running the SAR utility
during the performance measurement period. For Windows operating systems this
service could be provided by Performance Monitor snap-in for the Microsoft
Management Console. If you wish to supply Windows operating system OS
performance files, please ensure they are supplied in a text format (CSV) not
the default binary output from MMC logging. The output files need to contain the
"% Processor Time" output element from the Performance Monitor MMC Snap-In.
The performance monitoring scripts should be run for 7 (seven) clear days of
operation. This will ensure that we cover a full business week and a weekend of
operations.
Having run the performance monitoring scripts and O/S utilities for 7 days,
you can do one of the following :
either
Upload the log information into your own 9i database as defined in the Performance
Monitoring configuration guide, and then transmit an export file from
the database along with the OS monitoring logs.
Please ensure the export file contains the complete OWA_PERF user schema
and objects, not just the table data, the schema contains materialized views
for the data that will be useful during the reporting phase.
If you don't know how to run a DB export - you can read the
EXPORT documentation, or contact your DBA
or
Package up the logs on their own in a zip file and transmit them to Oracle
for processing.
We will load the logs into our own repository and analyze them in situ.
Checklist
When you have completed the SBE running period please ensure that you have
the following components before transmitting them to us
Exported OWA_PERF database dump file OR seven days of mid-tier log files
Seven days of OS monitoring data
Completed Portal Implementation Case Study : Customer Questionnaire
When you have these 3 components please email them to
portal-sbe_us@oracle.com. If you have
any problems please send an email with the details to the same address, and we
will try to resolve them for you.
Please do NOT contact Oracle Support services for assistance with this SBE
feedback program.
Bookmark
