Klaus Aschenbrenner
@sqlpassion.at
CEO & Founder of https://SQLpassion.at. SQL Server & PostgreSQL specialist during the day. Astrophotographer during the night. Triathlete in between. Loves low-level system programming in Rust.
Who would get the compensation? The author or the publisher?
November 19, 2025 at 12:08 AM
Who would get the compensation? The author or the publisher?
ZSTD with level MEDIUM gives you the same runtime as the traditional MS_XPRESS algorithm, but the backup is around 14 GB smaller. The level LOW is really fast (203 sec. vs. 305 sec., and the backups are still smaller (58 GB vs. 50 GB). #sqlserver #zstd #compression #performance
November 15, 2025 at 3:02 PM
ZSTD with level MEDIUM gives you the same runtime as the traditional MS_XPRESS algorithm, but the backup is around 14 GB smaller. The level LOW is really fast (203 sec. vs. 305 sec., and the backups are still smaller (58 GB vs. 50 GB). #sqlserver #zstd #compression #performance
A simple comparison across the various backup compression algorithms in SQL Server 2025 (230 GB database backup on a 8 vCore VM, 16 GB RAM on Proxmox):
MS_XPRESS: 305 sec => 58 GB
ZSTD LOW: 203 sec => 50 GB
ZSTD MEDIUM: 310 sec => 44 GB
ZSTD HIGH: 936 sec => 41 GB
MS_XPRESS: 305 sec => 58 GB
ZSTD LOW: 203 sec => 50 GB
ZSTD MEDIUM: 310 sec => 44 GB
ZSTD HIGH: 936 sec => 41 GB
November 15, 2025 at 3:02 PM
A simple comparison across the various backup compression algorithms in SQL Server 2025 (230 GB database backup on a 8 vCore VM, 16 GB RAM on Proxmox):
MS_XPRESS: 305 sec => 58 GB
ZSTD LOW: 203 sec => 50 GB
ZSTD MEDIUM: 310 sec => 44 GB
ZSTD HIGH: 936 sec => 41 GB
MS_XPRESS: 305 sec => 58 GB
ZSTD LOW: 203 sec => 50 GB
ZSTD MEDIUM: 310 sec => 44 GB
ZSTD HIGH: 936 sec => 41 GB
I also hope so. Am I allowed to demonstrate your example/approach (based on my own data) in my upcoming SQL 2025 online training?
November 14, 2025 at 9:30 PM
I also hope so. Am I allowed to demonstrate your example/approach (based on my own data) in my upcoming SQL 2025 online training?
I don't want to count it exactly, but #PostgreSQL has at least 50+ JSON-related functions... www.postgresql.org/docs/18/func...
9.16. JSON Functions and Operators
9.16. JSON Functions and Operators # 9.16.1. Processing and Creating JSON Data 9.16.2. The SQL/JSON Path Language 9.16.3. SQL/JSON Query Functions …
www.postgresql.org
November 14, 2025 at 5:53 PM
I don't want to count it exactly, but #PostgreSQL has at least 50+ JSON-related functions... www.postgresql.org/docs/18/func...
*) SQL Server 2025 introduces Optimistic Locking (TID Locking, LAQ) which is based on ADR.
PostgreSQL: Multi-Version Concurrency Control (MVCC) since its beginnings - and that's it.
Sorry, but this is NOT easy and understandable anymore!!! #SQLServer #PostgreSQL #postgres
PostgreSQL: Multi-Version Concurrency Control (MVCC) since its beginnings - and that's it.
Sorry, but this is NOT easy and understandable anymore!!! #SQLServer #PostgreSQL #postgres
November 13, 2025 at 4:38 PM
*) SQL Server 2025 introduces Optimistic Locking (TID Locking, LAQ) which is based on ADR.
PostgreSQL: Multi-Version Concurrency Control (MVCC) since its beginnings - and that's it.
Sorry, but this is NOT easy and understandable anymore!!! #SQLServer #PostgreSQL #postgres
PostgreSQL: Multi-Version Concurrency Control (MVCC) since its beginnings - and that's it.
Sorry, but this is NOT easy and understandable anymore!!! #SQLServer #PostgreSQL #postgres