Should We Defragment Indexes?
summary
Argues that external fragmentation can make defragmenting counterproductive: rebuilding or reordering pages often concentrates them onto a single capacity device or controller in a VM/hypervisor storage layout, reducing parallel I/O and hurting throughput; contrasts that with fixing internal fragmentation (rebuilds reduce page count and can help the optimizer). Uses simple card-demo and storage-layout reasoning rather than heavy tooling, recommends benchmarking index maintenance in each environment (Azure SQL DB/Managed Instance, VMs) and favors cautious, measured reorgs over blind rebuild schedules.
AI-written summary · about
chapters (8)
- 0:00Introduction to maintenance
- 1:43Internal vs external fragmentation
- 3:32Reorgs vs rebuilds
- 5:26Impact on storage architecture
- 8:56The controller analogy
- 9:44Conclusion of main talk
- 10:47Q&A session
- 19:04Closing remarks
related talks
19mShould We Defragment Indexes?
19mShould We Defragment Indexes?
53mIndexing Internals for Developers & DBAs
53mIndexing Internals for Developers & DBAs
53mIndexing Internals for Developers & DBAs
53mSQL Server Internals: The Practical Angle
11mDatabase Evolution: Partitioning and Indexing
47mArchitecting for High Performance SQL Server on Virtual Machines
