PSP optimization

SQLServerCentral Article

Exploring Parameter Sensitive Plan Optimization in SQL Server 2022

  • Article

PSPO (Parameter Sensitive Plan Optimization) is a SQL Server feature that improves query performance by accepting varied data sizes based on the runtime parameter value(s) specified by the customer. It deals with the situation in which a single cached plan for a parameterized query isn't the best option for all potential incoming parameter values. Non-uniform data distributions exhibit this phenomenon. When using PSPO, SQL Server keeps several execution plans for a single query, each one customized for a particular parameter value. With the help of this feature, numerous execution plans for a parameterized query are generated, each of which is tailored for a certain range of parameter values.

You rated this post out of 5. Change rating

2023-07-21

2,388 reads

Blogs

Part 2: The Modern Azure Data Warehouse – ADF and ADLS

By

In today’s data-driven world, having the right tools to manage and process large datasets...

A New Word: Flichtish

By

flichtish – adj. nervously aware how much of your self-image is based on untested...

Automating Container Image Builds with Docker Build Cloud and Github Actions

By

In a previous post we went through how to use Docker Build Cloud to...

Read the latest Blogs

Forums

Backup with Checksum Show No Issues, While DBCC CHECKDB Reports Errors

By IT researcher

I took a backup with the CHECKSUM option, which was successful and showed no...

Realities of Predictive Analysis

By Louis Davidson (@drsql)

Comments posted to this topic are about the item Realities of Predictive Analysis

Linked Server Product Name?

By Brandie Tarvin

Just came across something weird. Was wondering if anyone else has ever seen this....

Visit the forum

Question of the Day

Creating Statistics on XML

I have a table in SQL Server 2022 that is an XML data type. I have an index on this column and want to create statistics using CREATE STATISTICS for some of the data in the column. What should I do?

See possible answers