Help

Contents
Search

Application Audit Online Help
Compuware Enterprise Services
Fault Analytics
ISPW Web Interface
ThruPut Manager Web
Topaz For Java Performance
Topaz for Total Test Web Client User Guide
- Welcome to Topaz for Total Test
- Topaz for Total Test Non-virtualized Test Overview
- Getting Started
  - Getting Started Tutorial
  - Using the Web Client
  - Create a Test Suite from Browser
- Test Virtualization
  - Test Virtualization
  - Stubs in the Web Client
  - Recording Input Data to Sub Modules
- Extending Non-virtualized Test with Custom Data Definitions
- Code Coverage
- Configuring the Topaz for Total Test Repository Server
- Working with the Repository
  - Working with the Repository
  - General Information for All Artifacts
  - Defining and Using Components
  - Components in the Web Client
  - Connections
  - Environments
  - Systems
  - Categories
  - User Groups
  - Skins/Data and Execute
  - Stubs
  - Job Definitions
  - Dataset Skins
  - Test Scenarios
  - Suites
  - Reports
- Sample Programs
- Appendix A Tutorial Add a Component to the Repository with the Web Client
- Welcome to Topaz for Total Test
- Get a License
- Topaz for Total Test Functional Test Overview
- Getting Started
  - Getting Started Tutorial
  - Using the Web Client
  - Create a Test Suite from Browser
- Test Virtualization
  - Test Virtualization
  - Stubs in the Web Client
  - Recording Input Data to Sub Modules
- Extending Functional Test with Custom Data Definitions
  - Extending Functional Test with Custom Data Definitions
  - Extending a Test Scenario
  - Extending the Execution Context
- Code Coverage and IDz Integration
  - Code Coverage and IDz Integration
  - Browser Client
  - Eclipse Client
- Utility Tools
  - Utility Tools
  - Environment Specific Host UserIDs
  - Management Of Saved Host Credentials
  - Browse Datasets
  - Get Job Info and Spool Files
  - Copy And Delete Datasets
- Configuring the Topaz for Total Test Repository Server
  - Configuring the Topaz for Total Test Repository Server
- Working with the Repository
  - Working with the Repository
  - General Information for All Artifacts
  - Defining and Using Components
  - Components in the Web Client
  - Connections
  - Environments
  - Systems
  - Categories
  - User Groups
  - Skins/Data and Execute
  - Stubs
  - Job Definitions
  - Dataset Skins
  - Test Scenarios
  - Suites
  - Reports
  - Schedules
- Sample Programs
  - Sample Programs
  - Program XAPGMTST
  - Program XADB2TST
  - Program XAMOCTST
  - Program XAPGMCAL
  - Program XAPGMBT1
  - Program XARISCAL
  - Program XADTSTMP
  - Program XADGTIME
  - Program XADUTF8
iStrobe Online Help

Analyzing the Data Set Characteristics Report

If a file has a high EXCP count on either data or index portions, check the I/O Facility Usage report to get a further breakdown of access by unit and volume. If a file has CI or CA splits, consider increasing FREESPACE percentages.

If a sequential file has many EXCPs to its data portion, and few buffers, consider increasing data buffers or blocksize. If it also has high EXCPs to its index portion, the index CA size might be too small to accommodate sufficient pointers to the data. The SPACE parameter in the JCL controls index CA size. VSAM uses the smaller of the primary and secondary allocations (with a maximum of 1 cylinder) for the Index CA size.

If the access method for a data set is QSAM and the number of buffers is 5, you are using the default number of buffers for a QSAM file. This is normally not as efficient as is possible. Increase the number of buffers to 15 or 20 and remeasure. You should see a reduction in the number of physical I/Os and a decrease in the wait time for the file. The optimum number of buffers varies depending on your individual situation.

If the access method for a data set is VSAM INDX, and OPEN INTENT includes DYN or DIR, for optimum performance, ensure that you have allocated enough buffers to the index component. The minimum number of buffers you should allocate for an index component is the number of index levels plus one. You can determine the number of index levels by clicking the icon to the left of the file name.

If a randomly-accessed file has many EXCPs to its index portion, and few buffers, consider increasing index buffers.

For information on data set activity, review this and other reports that give data set information in addition to ddname information.

Because of the complexity of managing VSAM and DASD access, consult with your site's specialists on the results of your measurements and possible changes to your applications.