Re: Same query with different response time

Thank you Mladen. I checked the historical size of tablespaces(from dba_hist_tbspc_space_usage) and summed them up for this DEV database. And it comes as prior to TSE the database size was ~633GB and post TSE it's now ~636GB, so it's showing some increase but not showing a 2-3 times kind of increase at least.

But one thing I wanted to check , as this encryption technique is widely used one, is such degradation in performance expected (as because it's degrading from 50% to 100% for some queries)?  OR
Are we testing/validating performance here in the wrong way?

Regards
Lok

On Mon, Jun 28, 2021 at 9:33 AM Mladen Gogala <gogala.mladen_at_gmail.com> wrote:

> IO has changed because encrypted data is much larger than the original.
> Encrypted data can be 4 times or more larger than the original. That is
> what happens when you apply ciphers like RSA RC4. There are many
> decryption techniques that are being stymied by this: pattern
> recognition, same character repeat and many others. Since the keys are
> enormous, you would need 100 years or so to break them. Public key
> encryption is based on huge numbers that need to be factored into their
> prime components. The most efficient factorization algorithm is the
> Sieve of Erathosten, named after mathematician form the ancient Greece.
> So, unless some mathematical wunderkind comes up with something much
> faster, your data is safe and secure. That hasn't happened for the last
> 2500 years or so. The price you pay is much enlarged data size and a lot
> of CPU cycles to decrypt stuff. Personally, I use ROT13 and, to be safe,
> I encrypt everything - twice.
>
> On 6/27/21 10:18 PM, Lok P wrote:
> > No other change happened. It was only tablespace encryption which has
> > been done on top of existing data.
> >
> > I am also struggling to understand why the increase in IO is noted in
> > SQL monitor.
> >
> --
> Mladen Gogala
> Database Consultant
> Tel: (347) 321-1217
> https://dbwhisperer.wordpress.com
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>

--
http://www.freelists.org/webpage/oracle-l