Hello Guys,
Basically I am doin a basic course in Database Design Management System in my BS I have read about normalization but had a problem in it. I want U guys to help me I need some technical specification regarding this topic.

My Query is:
"Database normalization is the process of organizing the fields and tables of a relational database to minimize redundancy and dependency. but as it removes anomalies form data it effect query performance and why does this happen???"

Take care

OK...WHO is teaching this course

They are going to say that the normalized model will slow down data access do to the number of joins associated with retrieving the data

That may be, but a denormalized structure INCREASES the risk of having data corruptions...so the speed you may incur (MAY) is offset by data corruption

So, if the bozo's want to say that retrieving data from a single table than many is faster...fine..can't argue with that...well maybe you could...but at what COST?

Bozo's

And not for nothing...ALL of my sprocs are sub second...even in the thousands of seconds...I know.. because I log their execution

Now...if Professor Bozo wants you to retreive 50,000 rows for a report, then there is something wrong with Professor bozo

ALSO..it ONLY Affects performance if the tables are not properly Indexed

And retrieving data from a single table is not necessarily faster.
A denormalized dataset means a wider table, which means fewer records per page, which means more time required to search for individual pages. It also mean filtering by string-based values rather than more efficient numeric surrogate keys.
So, there are performance trade-offs either way, but the relational integrity of a normalized schema trumps all.

agreed

but a little disappointed you used a bridge metaphor instead of chess

mates all?

In general, as the data becomes more normallized performance improves. There are many factors that drive this performance increase, first of which is that normallized data takes less storage so more usable data can be held in faster storage than a non-normalized schema will allow. An important side effect is that queries against a normalized schema are statistically more likely to be logically correct so the user gets the answer on the first try instead of after multiple attempts and QA efforts.

-PatP

You and I BOTH know that this is NOT The answer the Prof is looking for, however CORRECT it is

It's a pervasive perverted assessment

http://www.lcard****/~nail/sybase/perf/1088.htm

The answer the Prof wants is

"It impove pe-formANCE...AH YUP....bobby jean...get me my corn cob pipe"

Normalization has nothing to do with performance because normalization defines only the logical features of the database (tables, attributes, keys, dependencies). Normalization says nothing about what the database looks like in physical storage or what optimizations can be applied to queries on it.

If and when you find that some schema change due to normalization affects query performance then that's because of insufficient physical database independence in the DBMS concerned. In other words the DBMS made some unwanted physical change even when you only required a logical change to the schema.

Can I get a hit?

This is what I am lookin for I think every one here is interpenetrating that it effect performance in a negative way but I personally think normalization increases performance Like only 1NF improves operations like SELECT and JOIN as search operations are reduced...I will really appreciate if some one will point me to the information It effect performance means It improves it as a byproduct and thats all...

My english is terrible so don mind it

They are going to say that querying 1 table is faster the querying 10

I'm not so sure that is case