Model Features

The Mainframe model suite implements the following features.

Table 151: Mainframe Model Features

Model Feature Description
Logical Partitioning Allows the machine to have up to 15 logical servers with each one of them running under Compatibility or Goal Mode.
CPU Modeling Multitasking CPU modeling uses a scheduling algorithm based on business importance of the workloads and their current performance index.
Memory Model tracks memory usage to help predict server memory requirements.
Disk I/O Currently a simple I/O subsystem is modeled.
Number of Physical zAAPs Includes IBM zSeries Application Assist processors in the model.

Mainframe models are implemented based on information available from the following sources.

Table 152: Reference Documents

Document Description
TSO (zSeries 900 2064–1C1) LSPR benchmarks published by IBM (www.ibm.com)
Manufacturer Specifications Number of CPs, interfaces, memory and SRM constant information are obtained from manufacturer data sheets.
  • A workload can be a single thread, a process, or a group of threads and processes that represent the target workload. Each workload has specific resource requirements for CPU, memory, and I/O.
  • Workloads can be executed as applications entirely within the mainframe, or as networked applications that scale with traffic load.
  • If operating under Goal Mode, workloads can be associated with Service Classes thereby specifying performance goals and resource groups for these workloads.
  • There are many predefined hardware configurations.
  • Quick, simple mainframe reconfiguration facilitates “what if” scenarios.
  • The model interoperates with existing “simple CPU” and “specialized Server” models.

Creating and Processing Jobs Using AppTransaction Xpert Whiteboard

Using the AppTransaction Xpert Whiteboard GUI or Python scripting API, you can create and process mainframe jobs.