Copy of 22 - PowerPoint PPT Presentation

1 / 22
About This Presentation
Title:

Copy of 22

Description:

Far and away, ERP fits into the cif as an. operational data store, ... becomes unmanageable. If there are m tables and n cubes, then there will be up to - m x n ... – PowerPoint PPT presentation

Number of Views:38
Avg rating:3.0/5.0
Slides: 23
Provided by: billi155
Category:

less

Transcript and Presenter's Notes

Title: Copy of 22


1
a presentation by W H Inmon
2
appls
ERP
There has been a movement to convert old
applications to ERP, primarily because of the Y2k
issue
3
ERP - - SAP - BAAN - PeopleSoft - JD
Edwards - Oracle Financials - Lawson -
Seibel - others
From a market share perspective, SAP is the
leading vendor
4
Corporate information factory
ERP
Far and away, ERP fits into the cif as
an operational data store, or an ODS or
as another source of data for the data warehouse
5
But getting information out of SAP is very
difficult to do
SAP
6
query
9,000 tables
SAP
The performance implications of trying to join
9,000 tables are outrageous
7
Documentation for the tables is cryptic
SAP
8
The internal structure of the tables is packed,
variable length The internal format is proprietary
SAP
9
No wonder there is so much frustration with SAP
when it comes to information management
SAP
10
SAPs BW
Now SAP offers up their version of a data
warehouse - BW
However, SAPs current release of BW is not a
data warehouse
SAP
11
What SAP has built is a bunch of
proprietary cubes on top of their application
12
As time passes, it is predictable that the
interface between the application tables and the
cubes becomes unmanageable
m
n
If there are m tables and n cubes, then there
will be up to - m x
n interfaces needed
13
27,298
36,559
45,110
35
1,220,991
982,338
Reconciliation between the different results
produced by the cubes is impossible
14
Trying to use third party tools against
the proprietary cubes created by SAP is
an impossibility
15
There is a massive amount of redundant data
16
Trying to change a single data value may cause a
massive number of cubes to be recalculated
17
Trying to mix non SAP data inside an SAP cube
requires putting the non SAP data inside SAP
Is that what you want to do?
18
- Oracle - DB2 - UDB - Informix - T/D - others
A much more rational approach is to pull SAP data
out of SAP and into a real data warehouse
19
a local or a global data warehouse
another alternative is to create an SAP local
data warehouse and relate it to other data
warehouses - local or global
a local data warehouse
20
In order to pull data out of SAP, a commercial
interface is the best bet - - Ardent -
SAS - D2k - Acta - others
21
Non SAP data can be easily integrated with SAP
data when a third party data warehouse is built
22
(No Transcript)
Write a Comment
User Comments (0)
About PowerShow.com