Re: GUID's and uniqueness

From: Bricklen <bricklen_at_zyahoo.zcomz>
Date: Wed, 15 Sep 2004 14:16:38 GMT
Message-ID: <a9Y1d.32271$XP3.11332@edtnps84>

Louis Frolio wrote:
> Daniel Morgan <damorgan_at_x.washington.edu> wrote in message news:<1095218401.235027_at_yasure>...
>

>>Louis Frolio wrote:
>>
>>
>>>Greetings All, I have read many upon many articles here regarding GUID
>>>data types and uniqueness.  There have been many opinions regarding
>>>the effectiveness of GUID's and when they should/should not be used. 
>>>However, every article strongly implies, if it does not state it
>>>outright, that GUID's are always unique.  My question is this, what
>>>happens if you have a database that uses GUID's and the NIC is changed
>>>out on the box?  From what I understand the MAC address of the NIC is
>>>used as part of the algorithm to generate a GUID.  If you change out
>>>the NIC after generating 1 billion GUID's do you run the chance of
>>>generating a duplicate GUID?
>>>
>>>I look forward to your insightfulness on this issue.
>>>
>>>Regards, Louis.
>>
>>GUIDs are not unique except in very specific controlled situations.
>>They are highly likely to be unique ... but there are not guarantees.
>>And I have personally seen problems where they weren't.

>
>
>
> Daniele, would be so kind as to elaborate a bit with regards to the
> situation where duplicate GUID's appeared? I ask because I am
> designing an application that requires that it be replication
> friendly, and not just one way replication. I need the app to be able
> to do two way replciation to multiple nodes. So, any details you can
> provide regarding your experience with GUID's would be hugely helpful
> to me.
>
> Regards, Louis.

Not sure if it will help, but I ran some extensive tests back in January on sys_guid() and regular sequences for insert and query speeds and contention etc. Sequences won hands down in almost all scenarios, IIRC. I couldn't find my notes, or I'd post some numbers.

Anyways, if your intent is more to have unique pk's across databases, an alternative is to have sequence numbers that start at different numbers, but increment at the same rate. Tom Kyte talks about this, as do several of the luminaries of this newsgroup.
link to get you started:
http://asktom.oracle.com/pls/ask/f?p=4950:8:4007837615079353646::NO::F4950_P8_DISPLAYID,F4950_P8_CRITERIA:7010972506678,

hth Received on Wed Sep 15 2004 - 09:16:38 CDT