|
|
Posted by mittal.pradeep on 09/21/05 19:00
1. But the data collection and reporting is in horizontal format. If
collected data is edited vertically, won't there be a extra steps of
converting horizontally obtained data to vertical and then vertical to
horizontal reports. In custom column model data always remains
horizontal. Won;t performance not be a issue in taking care of two
extra steps required in vertical model.
2. Won;t the concurrency be a issue, considering the fact that a
logical single horizontal row is edited as say 10 rows. Two people
might be changing same Primary key's different attributes at the same
time.
Erland Sommarskog wrote:
> (mittal.pradeep@gmail.com) writes:
> > What is the better table design for a data collection application.
> > 1. Vertical model (pk, attributeName, AttributeValue)
> > 2. Custom columns (pk, custom1, custom2, custom3...custom50)
> >
> > Since the data elements collected may change year over year, which
> > model better takes of this column dynamicness
>
> The vertical model is certainly cleaner from a relational perspective.
> It also requires less maintenance.
>
> But admittedly queries can be more complex. If attributes can be of
> different data types, you need some triggers to check this. A tip
> is that the sql_variant data type is good in this case.
>
>
> --
> Erland Sommarskog, SQL Server MVP, esquel@sommarskog.se
>
> Books Online for SQL Server SP3 at
> http://www.microsoft.com/sql/techinfo/productdoc/2000/books.asp
Navigation:
[Reply to this message]
|