Seeing the Forest AND the Trees: Capacity Planning for a Large Number of Servers

摘要

This author's capacity planning learning expertise was built on a limited number of servers. Specifically, it involved analyzing and reporting workloads on a few mainframe footprints and, eventually, a few dozen distributed (Unix and NT) servers. Today's environments are growing in complexity and may include hundreds of distributed servers along with larger mainframe footprints. The challenge in this increasingly common environment is to manage the complexity of capacity planning and reporting for this explosion in the numbers and types of data processing platforms. Although this discussion is oriented to large companies (with a lot of "trees"), these approaches can be used by large and small companies alike.