The Most Common Database Design Mistakes - Plan? What Plan?

 
 
By Brian Prince  |  Posted 2008-03-19
 
 
 

The Most Common Database Design Mistakes

By Brian Prince

Title Slide

The Most Common Database Design Mistakes

The Most Common Database Design Mistakes - Plan? What Plan?

Slide 1

Lack of planning was an oft-repeated theme. Poor planning is the fountain from which many other problems spring-

The Most Common Database Design Mistakes - Plan? What Plan?

The Most Common Database Design Mistakes - Whats My Name?

Slide 2

The failure to adopt a naming standard was also cited as a problem, with each developer putting his or her own meaning on column names. Naming standards are a simple solution, with each table name and column name having a common meaning. For example, use column names that end with _date for date columns and _code for coded value columns.


The Most Common Database Design Mistakes - Whats My Name?

The Most Common Database Design Mistakes - Too Much, Too Little, Too Late

Slide 3

Experts say that developers often use too many or too few indexes, both of which are bad for performance. The trick is to measure your performance: A simple query analysis often can tell you what you need to know to decide how many indexes are needed; other times, you may have to examine the overall server performance and how it's affected by the locking of specific indexes under load.


The Most Common Database Design Mistakes - Too Much, Too Little, Too Late

The Most Common Database Design Mistakes - Everybody Needs a Friend

Slide 4

When stakeholders aren't involved from the beginning, there can be problems. Including, for example, the SAN (storage-area network) and systems administrators in the planning stage can pay off in the end.


The Most Common Database Design Mistakes - Everybody Needs a Friend

The Most Common Database Design Mistakes - Tooling Around

Slide 5

Not using tools or features meant to guarantee data integrity is a mistake. Databases now include all kinds of tools meant to make your life easier, so there's no excuse for not having a thorough understanding of tuning and optimization features.


The Most Common Database Design Mistakes - Tooling Around

The Most Common Database Design Mistakes - Just What Is Normal?

Slide 6

When it comes to OLTP, poor normalization can hurt data integrity and performance. Be cognizant of the 3NF (third normal form), and use model validation tools to detect normalization issues. A technical walk-through of the model by someone other than the modeler helps, too.


The Most Common Database Design Mistakes - Just What Is Normal?

The Most Common Database Design Mistakes - This Is a RAID

Slide 7

Get to know RAID (Redundant Arrays of Inexpensive Disks). Develop a thorough understanding of RAID-5 versus RAID-1, and avoid RAID-5 when you are dealing with large-volume random access data patterns.


The Most Common Database Design Mistakes - This Is a RAID

The Most Common Database Design Mistakes - Lets Get Physical

Slide 8

It's a mistake to ignore the physical side of the data model -

The Most Common Database Design Mistakes - Lets Get Physical

The Most Common Database Design Mistakes - See More Slideshows Like This One

The Most Common Database Design Mistakes - See More Slideshows Like This One

Rocket Fuel