As a SharePoint farm grows beyond 80 million indexed items, co-locating index, crawl, and content processing components on the same servers creates I/O contention and crawl queue bottlenecks. This post covers the large farm reference architecture — dedicated index servers, distributed index partitions, and isolated crawl databases — and how to implement it with PowerShell.
Large Farm
Designing the Right SharePoint Search Topology for Production SPSE Farms
The default SharePoint search topology gets you up and running quickly. For production farms, deliberate design choices around component placement, high availability, and crawl isolation are what turn a functional search deployment into a resilient one. This post explains all six search components, when to use a single SSA vs. dual SSA, and how to design the right topology before running a single script.