The Cost of GUIDs as Primary Keys

By Bill Graziano on 8 January 2005 | Tags: Data Types , Indexes


Kristen submitted "In this article, Jimmy Nilsson presents the pros and cons of using globally unique identifiers (GUIDs) as the datatype for primary keys in SQL Server 2000. In doing so, he shows you test results that hint of performance characteristics and introduces you to a special type of GUID that he invented, called COMBs, that solves what otherwise might give you a big throughput problem.

He also point out that GUIDs are not guaranteed to be 'Globally Unique'"

Link: The Cost of GUIDs as Primary Keys


Related Articles

Using the TIME data type in SQL Server 2008 (6 March 2008)

Using Included Columns in SQL Server 2005 (14 January 2008)

Using the DATE data type in SQL Server 2008 (6 December 2007)

SQL Server Indexes: The Basics (26 November 2007)

Working with Time Spans and Durations in SQL Server (15 October 2007)

DATEDIFF Function Demystified (20 March 2007)

Using Indexed Computed Columns to Improve Performance (16 January 2007)

Microsoft SQL Server 2000 Index Defragmentation Best Practices (2 May 2004)

Other Recent Forum Posts

SQL Server detected a logical consistency-based I/O error: torn page (expected signature: 0xaaaaaaaa; actual signature: 0xfffffffe). It occurred during a read of page (1:908) in database ID 99 at offset (113m)

Case stmt in SSIS (19h)

Upgrading from MSSQL 2016 to 2017 (1d)

Why does CASE WHEN NULL IN (SELECT A FROM B) THEN 'Y' ELSE '' AS [NULL] does not work? (1d)

Tensorflow package installation on SQL 2019 ML Services (1d)

How can I reference the table (T1 JOIN T2 ON A=B JOIN T3 ON C=D)? (1d)

Is there an easy way to populate a column with an INDEX/MATCH statement? (2d)

Is joining table completely deceiving? (2d)

- Advertisement -