Build real-time operational analytics with a filtered columnstore index, inspect batch-mode plans and rowgroups, and measure write costs.
Median on Big Tables: PERCENTILE_CONT vs a Two-Row Seek
Two ways to get a median, one tiny demo, and the traps in NULLs, even counts and empty tables. Then measure on your own server.
Storing JSON in SQL Server
Storing JSON in SQL Server can use validated nvarchar or the native 2025 json type, with indexed paths and relational columns for stable facts.
Trigram Matching: Finding Similar Words With an N-Gram Table
Build trigram matching with GENERATE_SERIES and an indexed n-gram table, rank shared pieces, and compare spelling similarity with SOUNDEX.
Soft Delete With an IsDeleted Column: Indexes, Views and Keys
Design soft delete with an IsDeleted flag, filtered unique keys and indexes, a live-row view, and a measured archive process.
Narrow Index Keys: Why the Clustered Key Width Matters Everywhere
Evaluate narrow index keys by comparing clustered GUID and integer designs, measuring index pages, and preserving natural uniqueness.
An Index Seek That Reads Millions of Rows: Residual Predicates
Diagnose an index seek that reads far more rows than it returns by checking residual predicates, Actual Rows Read and index key order.