Hello

You should always use explicit cursors (CURSOR <cursor_name> IS) and avoid implicit cursors (SELECT INTO) entirely, because explicit cursors are always more efficient.
PLSQL: Controlling Mythological Code



The short answer is that implicit cursors are faster
Implicit vs. Explicit Cursors which is faster - cursor_comparison.sql script

I tested and satisfied myself that implicit cursors are better than explicit cursors
http://asktom.oracle.com/pls/asktom/...:1205168148688

I m confused,which one is faster?

Beyond these I have another questions ,suppose I export data
-I can run sql in toad and export it
-I can develop plsql procedure and export it
which one is faster

Best Regards

I didn't test it myself, but if I had to choose between Donald Burleson (your second link) and Tom Kyte (your last link), I'd rather believe what Tom says.

You need to read for comprehension, rather than for sound bites.

The first article is one debunking some oracle myths.

If you include the line above the one you quoted, then you get this quote: Already, it starts to sound less like a piece of advice.

If we include the paragraph following yours, we see that you've got the point of the article exactly wrong:

This is a perfect case/question that Tom Kyte preaches constantly:
TEST FOR YOURSELF! DO NOT RELY ON OTHERS INFORMATION.
BENCHMARK!

I would not rely on Burleson for anything reliable. He tests little and his information is frequently incorrect. All he does is cut/paste from the oracle docs (if that).

THINK about what a cursor is and what it does.
THINK about how oracle processes a cursor.

If you know how oracle works and you benchmark and test then you will have your answer for yourself and will have the documentation to prove it to anyone in your organization.

This reminds me of a co-worker who said:
running insert,update,delete statements are way faster than MERGE.

I said: What makes you say that?

He said: I read it. It's a known fact.

I said: PROVE IT. Benchmark this "known fact".

Guess what we found out?
Your best answer will be one you test.
MAINLY because someone will always ask you: Why did you implement it like that?
You can then say: I tested and benchmarked it. I chose the best option based on my findings. Let me forward you my results.