Subject: | |
From: | |
Reply To: | |
Date: | Fri, 28 Apr 2000 07:36:43 -0500 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
At 03:23 PM 4/27/2000 -0600, Timothy Hoefner wrote:
>I'm going at it again with our DBA but I can swear that I learned
>somewhere that TurboImage can hash faster with an X item
>rather than an I item.
As long as the key size is less than or equal to 4 bytes, the performance
can essentially be considered equal. However, when the non-binary key type
size goes beyond 4 bytes its rules involve folding-in the other bytes
whereas the binary calculation simply uses only the right-most 32-bit
value. From that perspective, the non-binary hash address calculation
for sizes > 4 bytes require more processing steps to derive the
primary hash address.
In terms of entry storage/retrieval, its a case of 'it depends'. From
a general view point, the non-binary key 'usually' has less collisions
and therefore, possibly less secondaries. But there are always exceptions.
I would strongly recommend that you go to the Adager web site and get a
copy of Fred White's papers on this subject to better understand how
this process works from a data value perspective. The papers are entitled:
"The Three Bears of IMAGE"
"The Use and Abuse of Non-hashing Keys in IMAGE"
"Integer Keys: The Final Chapter"
These are located at:
http://www.adager.com/TechnicalPapers.html
/jf
_\\///_
(' o-o ')
___________________________ooOo_( )_OOoo____________________________________
Thursday, April 20th
Today in 1949 - Cortisone was discovered.
___________________________________Oooo_____________________________________
oooO ( )
( ) ) /
\ ( (_/
\_)
|
|
|