[PLUG-JOBS] MySQL Gig

[Vaughn G. Frost]

Mr. Decoste,

 

I realize your post was from December of 2007; however, I felt you may be
interested in my input.

 

I am Vaughn Frost - over 30 years of I.T. experience.  My background spans
multiple companies and multiple engagements both in the United States and
around the world.  Currently, I work from home for a fortune 500 company
performing data architecture services.  

 

To attempt to identify where I would match your needs (based upon your terse
statement of need), I perform Business Analysis to glean logical data
structures.  I do this via the Erwin data modeling toolset.  Once the key
logical concepts are determined and validated (to the business
rules/requirements), I create the MySQL design based upon those business
rules and other underlying data architecture concepts.  With that completed,
the process of designing the application begins.  The concept is to follow
the DBMS structures with the application code to create a system that
encompasses all current business needs with an eye on growth potential.
That potential being available in a DBMS that is properly designed to
accommodate such growth without costly application code re-writes.

 

As for 'MySQL Guru', I have implemented some internal business MySQL
solutions over the past several years - of which the most recent was for a
manufacturing process.  The MySQL version used was 4.1 using multiple
engines (INNODB and MYISAM).  I elected to not use referential integrity (FK
constraints); however, the system could use them with no issues.  Some of
the key elements I looked for and designed were:

1)       Separation of information into 3NF (3rd Normal Form) so that
repetition and duplication is handled logically.

2)       Creation of underlying code/reference tables that drive attributes
of the system design (e.g. Color Codes for electrical wiring)

3)       Use of link names for large objects (e.g. *.jpg files) such that
MySQL queries can locate the jpg to be displayed without burdening the MySQL
database with the large object item itself.  That is, the jpg files are
actually stored in a separate folder on the Linux file system.

4)       Allow for dynamic upload of new images with automatic naming.

5)       Architecturally:

a.       Design reference/code tables (at a minimum) to allow for deletion
'flags' instead of physical delete; therefore, allowing the application to
utilize old code values on display but not for new instances of code usage.


b.       Limit the number of rows in tables for key application data needs
by creating 'inactive' prefixed tables and logic to offload
unused/inactivated information on an ongoing basis (part of application
design)

6)       Create indexing schemes for each table to accommodate complex MySQL
queries (joins and where clause use).

7)       Develop a backup/recovery plan (Data Continuity) to minimize (or
eliminate) any potential data loss from any hardware/network issue. 

 

 

 

Vaughn G. Frost
Data/Information Architect

vaughn at service4gis.com 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.pdxlinux.org/pipermail/plug-jobs/attachments/20080602/68f07703/attachment.html>