Re: Is this a good use for an IOT type table or are there better solutions.

From: Daniel Morgan <damorgan_at_x.washington.edu>
Date: Wed, 24 Mar 2004 11:29:59 -0800
Message-ID: <1080156575.311711@yasure>

colin_lyse wrote:

> In article <1080113669.109456_at_yasure>, Daniel Morgan <damorgan_at_x.washington.edu> wrote:
>

>>Are the ID numbers sequential growing in a manner where all inserts 
>>would be at the 'end' of the table and not be into existing blocks?
>>Are there ever updates or deletes?

>
>
> no, id numbers aer pretty static grow by 2-3% a year at most. so inserts would
> be into existing blocks.
>
> No updates or deletes, until we start pruning in around 3-4 years, and in that
> case a single delete, probably recreate the table at that point.
>
>

>>But the question on my mind is why? This table is so small that space
>>the little bit of space used by the index is inconsequential. What is
>>the business problem you are attempting to solve. Also the number of
>>inserts is very small ... only 70 per minute on average. Hardly seems
>>worth even thinking about.
>>

>
>
> Cost, Costs, Costs. This is just one server out of 100s and not the biggest
> by far (we have on another a 2 billion row table that grows by 4 million a
> day). However, even the added cost of the 2-3 gig storage is frowned upon by
> management, 'can we get by with less space?', 'is there a way to cut the
> cost'?, 'budgets are frozen, no money left for even 1 more byte', etc. So we
> are always looking at cutting costs. Since we are a tiny fraction of much
> larger group, our say 2 gig saving multiplied 1000 times, can start to add up.

Then I'm with Jonathan and would seriously consider an IOT. My only concern is that inserts into existing blocks carries a cost in an IOT but given the low transaction volume of this table ... pretty much so what.

With your other tables it might become more of an issue.

-- 
Daniel Morgan
http://www.outreach.washington.edu/ext/certificates/oad/oad_crs.asp
http://www.outreach.washington.edu/ext/certificates/aoa/aoa_crs.asp
damorgan_at_x.washington.edu
(replace 'x' with a 'u' to reply)

Received on Wed Mar 24 2004 - 13:29:59 CST