SQL Server Performance Blog where milliseconds matter

SQL Server 2025 Regex Support: Native T-SQL Expressions SQL Server 2025 && Regex = Less Code, Less Pain in your queries Hi Guys, Here’s some good news: after years of workarounds, hacks, and tricks, SQL Server 2025 finally ships with native regex support in T-SQL. And no, this isn’t just a nice-to-have. It’s a game changer for anyone who deals with messy data, phone numbers, emails, codes… you name it. For years we’ve had to rely on ugly patterns like LIKE '%...%' , CHARINDEX , SPLIT , or even external CLR functions. They worked, but the cost in complexity, performance, and bugs was too high. Finally: native regex in SQL Server 2025 That’s right. Now you get built-in functions such as: REGEXP_LIKE(...) REGEXP_REPLACE(...) REGEXP_SUBSTR(...) REGEXP_COUNT , REGEXP_SPLIT_TO_TABLE , REGEXP_INSTR , REGEXP_MATCHES (table-valued functions are rolling out too) All powered by the RE2 engine : fast, familiar, and compatible enough with ...

SQL Server 2025: Faster Backups with ZSTD Compression Hi Guys! One of the most practical improvements in SQL Server 2025 —and a very welcome one for DBAs constantly fighting backup windows—is the new support for Zstandard (ZSTD) compression. This modern compression algorithm delivers faster backups without significantly increasing CPU or sacrificing storage savings. What’s New Compared to Previous Versions Up to SQL Server 2022, backup compression relied on MS_XPRESS , with optional acceleration via Intel QAT (or software fallback using QATzip). But now with SQL Server 2025, Microsoft introduces ZSTD : an open-source, high-speed compression algorithm with flexible levels and excellent performance-to-compression ratios. How to Use It No new UI, no new tools. It’s all in T-SQL: BACKUP DATABASE YourDB TO DISK = 'YourPath\backup.bak' WITH COMPRESSION (ALGORITHM = ZSTD, LEVEL = LOW|MEDIUM|HIGH); LOW is the default and fastest MEDIUM ...