Mike,

I wish I'd made note of the custom autonum routine that Bill 
mentioned, but Dennis McGrath tweaked it a bit on the list so he 
could probably fill you in.  If you were going to have separate tables 
I assume you would have had something worth indexing in each. 
Combined with an integer column to identify the set of data you'ld 
be working with I would think you'ld be OK.... <famous last words>

I think you'ld be _way_ (read _way_ <g>) ahead to spend time 
optimizing your structure and indexing scheme as opposed to what 
it would take to manage 700 tables, let alone 700 separate 
databases.

> and the only useful index would have at least 700 duplicates

I'm not sure what you mean??

Sounds like a fun project

Ben Petersen


On 26 Sep 2001, at 8:19, Michael Young wrote:

> Hi Ben,
> 
> I thought about this but the table would end up with millions of rows of data 
> and the only useful index would have at least 700 duplicates and I am not 
> sure how efficient that would be. It would also introduce another data 
> problem in that there will have to be what is essentially an "autonumber" 
> going on for each of the 700 tables in one column. This is not a true 
> autonumber but can be done with the autonumber command if each is in it's 
> own table. The number in this column must begin with 1 and be sequential 
> based on the index column mentioned above. Then this column will have an 
> index built on it which will give another 700 duplicates in the index. Most of 
> the work done on the table will involve this "autonumber" column.
> 
> Speed is going to be important. Every day there are around 120 complex 
> calculations made for each table with the results being stored in the table. The 
> report analysis is a whole other very complex operation.
> 
> I guess at this point I could ask what is the maximum number of rows that a 
> table can accept and at what point does it become too unwieldy.?
> 
> Best,
> Mike Young
> 
> On Tue, 25 Sep 2001 23:30:44 +0100, Ben Petersen wrote:
> 
> >Mike,
> >
> >If these tables have identical structures, couldn't you add one more 
> >integer column to ID which ..data set.. this belongs to (1-700) and 
> >manage from there? Just a thought.
> >
> >Ben Petersen
> >
> >
> >
> >On 25 Sep 2001, at 23:00, Michael Young wrote:
> >
> >> Hi Troy,
> >> 
> >> I might be better off putting it into a few R:Base databases than DBASE 
> files, 
> >> but only testing will tell. I thought the same thing about this and I have 
> >> wracked my brain to get around this many tables but it appears there is 
> none. 
> >> The only way to cut down the number of tables is if we could create a 3 
> >> dimensional table and I don't think that is slated for the next version of 
> >> R:Base <G>.
> >> 
> >> Best regards,
> >> Mike Young
> >> 
> >> On Tue, 25 Sep 2001 21:34:33 -0600, Troy Sosamon wrote:
> >> 
> >> >Speed wise using any foreign data source is a lot slower than native 
> R:base, 
> >> >even using the sconnect between R:base database is slow.
> >> >
> >> >700 tables w/ 130 col in each.  I think the relational database police will 
> >> >come hunt you down???
> >> >
> >> >Troy
> >> >
> >> >>===== Original Message From [EMAIL PROTECTED] =====
> >> >>Thanks Manuel and Castanaro,
> >> >>
> >> >>Looks like I will probably end up going the DBASE route.
> >> >>
> >> >>Best regards,
> >> >>Mike Young
> >> >>
> >> >>On Tue, 25 Sep 2001 17:21:39 -0700, Manuel de Aguiar wrote:
> >> >>
> >> >>>Hello Mike,
> >> >>>----------------------------------------------------------------------------
> >> >-----
> >> >>>Tables/Views/Columns/Rows
> >> >>>
> >> >>>Create up to 16,000 tables or views
> >> >>>Create up to 16,000 columns
> >> >>>Create up to 400 columns (fields) per table or view
> >> >>>Maximum row (record) lengths of 4096 bytes
> >> >>>----------------------------------------------------------------------------
> >> >--------
> >> >>>I found this information at:
> >> >>>                    http://rbase2000.com/Products/rbase.htm
> >> >>>Hope this helps,
> >> >>>Manuel
> >> >>>
> >> >>>Michael Young wrote:
> >> >>>
> >> >>>> Hello,
> >> >>>>
> >> >>>> Does anybody know the limits in terms of number of tables and 
> >> columns
> >> >>that
> >> >>>> are allowed in RBWIN 6.5++?
> >> >>>>
> >> >>>> I am looking at developing a database with approximately 700 
> tables 
> >> and
> >> >>around
> >> >>>> 130 columns in each table.
> >> >>>>
> >> >>>> I realize that I may need to use multiple databases but I have 
> another
> >> >>option and
> >> >>>> that would be to create DBASE tables since there really is no 
> >> relationship
> >> >>>> between most of the tables and I can just attach them as needed. 
> Does
> >> >>anyone
> >> >>>> know the speed or other drawbacks to keeping data in a DBASE 
> file?
> >> >>>>
> >> >>>> Thanks,
> >> >>>> Mike Young
> >> >>>
> >> >
> >> >Troy Sosamon
> >> >Denver Co
> >> >[EMAIL PROTECTED]
> >> >
> >> 
> >> 
> >> 
> >> 
> >
> >
> 
> 
> 
> 


Reply via email to