Performance

Performance recipes are about keeping expensive language services responsive as the workspace grows.

In Pegium, the first performance tool to reach for is usually caching.

Typical cases

• computing document-local symbol tables
• memoizing derived semantic data
• reusing per-type index views
• avoiding repeated whole-workspace scans

Start with Caches, then return to the Document Lifecycle page if you need to reason about when cached data becomes stale.