If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

 
Go Back  dBforums > General > Database Concepts & Design > designing DB for the finance industry

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
  #1 (permalink)  
Old 03-04-04, 23:29
cybershadow_jp cybershadow_jp is offline
Registered User
 
Join Date: Dec 2003
Location: Tokyo Japan
Posts: 26
Question designing DB for the finance industry

hi to all,

i have been designing databases now for about a couple of years... however, they were all "supply chain"

this year, i would be responsible for designing DBs for the finance industry

i do know that the old and test rules of DB design could be applied - however, there are specific characteristics about the finance industry that just makes it difficult for me to apply (like lots of ATTRIBUTES, and LONG ones at that!)

i was kind of wondering if anyone could lead me to a certain reference site, or at least suggest a book that would allow me to at least make better designs for the finance industry (life insurance, investments, etc.)

thanks in advance!!
Reply With Quote
  #2 (permalink)  
Old 03-05-04, 09:55
Pat Phelan Pat Phelan is offline
Resident Curmudgeon
 
Join Date: Feb 2004
Location: In front of the computer
Posts: 12,605
Working in that industry, I can say that that in any group of N people, there will be approximately N * Log(N) opinions on how a database shoudl be structured.

When you work in the financial industry, you have to clearly separate OLTP from OLAP usage. The two are very different beasts.

OLTP you need to normalize until you can't normalize it anymore. Third normal form is a good start. Anything less and you'll go insane before product rollout. More is better.

OLAP needs to be designed for simplicity. The financial guys are rocket scientists in the financial arena, but often can't find the door in a relational world. They tend to think in (wildly complicated) spreadsheets, not in tables.

There are at least two books that I know of in the works, but neither one is even near being published. They'll shed a great deal of light on these issues, from a real world perspective. As of right now though, I don't know of any single reference that will help much.

-PatP
Reply With Quote
  #3 (permalink)  
Old 03-07-04, 19:17
cybershadow_jp cybershadow_jp is offline
Registered User
 
Join Date: Dec 2003
Location: Tokyo Japan
Posts: 26
Quote:
Originally posted by Pat Phelan

Working in that industry, I can say that that in any group of N people, there will be approximately N * Log(N) opinions on how a database shoudl be structured.

When you work in the financial industry, you have to clearly separate OLTP from OLAP usage. The two are very different beasts.

OLTP you need to normalize until you can't normalize it anymore. Third normal form is a good start. Anything less and you'll go insane before product rollout. More is better.

yes...

since i have designed a couple of data warehouses before (for the dessert/beverage industry), this wouldnt be a problem

Quote:


OLAP needs to be designed for simplicity. The financial guys are rocket scientists in the financial arena, but often can't find the door in a relational world. They tend to think in (wildly complicated) spreadsheets, not in tables.

hhhmmm... i think that summarizes it quite neatly... as of this weekend i did check out the current database - and wow!!! each table runs for at least 20 columns (based on experience, tables with more than 20 columns need to go through further normalization). well, it could be an OLAP system, but when i dug in further, this particular table is for online transactions wherein client policies are entered, say thousand of times a day

unfortunately, most of the design is based on a third-party system called INGENIUM by SOLCORP - is anyone familiar with this?... it seems that this was originally based on a COBOL design, and then ported to DB2 (our production system uses Oracle as the back-end)

Quote:


There are at least two books that I know of in the works, but neither one is even near being published. They'll shed a great deal of light on these issues, from a real world perspective. As of right now though, I don't know of any single reference that will help much.

-PatP
is that so... *bummer*
but i would appreciate any kinds of links - my greatest worry right now is how to convince these financial geeks how to follow database modelling rules and/or when to decide between following traditional design rules, or just go with the flow (ie., the INGENIUM)

thanks
Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On