Introduction to Services
While it is easy to estimate the potential impacts of new processors, faster disks, or new tape technologies on an individual job, predicting the impact that one or more of these changes might have on your batch or job completion window is a complex operations research problem. Performance Associates, Inc. has the tool set and experience to help your enterprise answer these questions.
When data movement between zOS and open systems platforms are critical for an end-to-end process, the problem of measuring and predicting performance becomes even more complex.
The Data Movement Audit
Dr. H. Pat Artis and his team at Performance Associates, Inc. developed Benchmark Methodology that measures the processing path lengths (CPU instructions per unit of transferred data) within a zOS Operating System for file transfer operations. The result provides a constant measure of the relationship between CPU cycles consumed and the amount of data transferred.
This approach and its results create an empirical foundation for measuring and predicting utilization and throughput associated with data movement between systems. Alebra Technologies, Inc. white paper “Efficiency and Performance - Comparing PDM to FTP and NFS” describes this approach, its measurements, and comparative results.
With this methodology as a foundation, Performance Associates, Inc. can analyze SMF data for specific intervals, providing a profile of the data movement activity associated with workloads and estimate their consumed resources.
The Deliverable
A summary report and graphical representation of the amount of data moved in each fixed time interval will provide a profile of current workloads associated with data movement. An estimate of processor utilization and throughput associated with those identified workloads will be provided. This approach can be used as a basis for planning resources for anticipated future workloads, measuring impact of configuration changes on job completion times or estimating future processor, network or storage requirements.
Alebra recommends that this Data Movement Audit be repeated at periodic intervals to prove the return on investment from investments in IT assets and to predict future needs.
