Best Practices for Mainframe I/O SLA and Efficiency Optimization

Features for the z/OS storage infrastructure such as FICON, Multiple Allegience, and Dynamic PAV have all shifted the I/O queuing bottlenecks from the host to components inside the storage systems.   Consequently, these bottlenecks have created a visibility gap into this mission-critical part of the IT infrastructure.   The ramifications of this visibility gap are felt both in the efficiency of the infrastructure and in SLA levels.   Performance problems asociated with storage for example, are typically not realized until after they already have affected production workloads.   In some cases, poorly performing storage hardware is replaced prematurely.  When performance wins are needed in batch or online jobs, the impact of faster I/O (vs. CPU) is important to consider.   This session will describe the architectural changes and which metrics and methodologies are necessary to reduce SLA risks and improve efficiency.

Tracks: Data Center and IT Management