세션 2-2(허태경).ppt

Size: px

Start display at page:

Download "세션 2-2(허태경).ppt"

이현 예
5 years ago
Views:

1 , an IBM Company 2005 IBM Corporation

3 Discover Prepare Transform & Deliver????????? Time To Value

TRANSFORM and DELIVER DataStage Service-Oriented Architecture,,

4 DISCOVER ProfileStage Service-Oriented Architecture Event Management PREPARE,, QualityStage Enterprise Connectivity TRANSFORM and DELIVER DataStage Service-Oriented Architecture,, Connectivity IT (XML, EDI, JMS, JCA) Complementary To BPM, EAI, and EII Technologies

5 WebSphere Data Integration Suite,, ROI Discover Prepare Transform Ascential Enterprise Integration Suite Time-to-Value Impact on Strategic Objectives

6 Custom Apps on Mainframe Data Requirements Data Extraction Criteria Functional / Technical Input Template Common Target Format Custom Apps on UNIX 3 rd Party SW SAP Oracle Siebel PeopleSoft Initial Extraction Extract & Load into Stage 1 Data Assessment Discover Assess & Validate individual sources Data Alignment Cleanse Standardize individual sources Data Harmonization Consolidate Integrate Cleanse Normalize & Harmonize across sources Solution Implementation Target System Prepare Transform Load Load OLTP SAP BW SAP R/3 EDW External Data DataStage ProfileStage & DataStage QualityStage & DataStage QualityStage DataStage

7 Measure Analyze Improve

12 :, Training or Skill Why so many project overruns or failures? :,

13 rofilestage! /,,. - Doug Laney (4/3/2002)

14 hat does ProfileStage do? Interactive analysis of all data sources Table & Primary Key Analysis / Column Analysis Source 1 Source 2 Cross-Table & Relationship Analysis Provides comprehensive understanding of the data,,,,, to & job &

15 inimize Project Resources and Risk Projects with extensive manual source analysis Source System Analysis Transform & Risk!!! Projects with limited manual source analysis Analysis Transform Test Deploy!!! Projects using ProfileStage s automated data profiling Analysis Transform Test Deploy No surprises!! Analysis OR Project cancelled:!!!

26 ualitystage Data Harmonization QualityStage, Data Re-Engineering Investigation ( ) Standardization ( ) Matching ( ) Survivorship ( ) Assessment Customers Transaction Localization Rule Set Materials Vendor/ Supplier 1. Investigation 2. Standardization 3. Matching 4. Survivorship DB

27 ata Preparation Data Re-Engineering Step 1) Investigation bbbbbbbbbbbb nnnbnnnbnnnn nnnbnnnnnnnn Percentage % % 7.810% nnnbnnbnnnnb 2 Percentage C_SG_BK % C_ST_BK % C_B_BK % C_ST_DG_BK % C_SG_SK % % ^J,^?^?^B,?,^?,^?^?^E,N % 0.020% TEETH,1-3/4" BORE,TAPERLOCK,50 PITCH,3-5/16" OD,D Rule Set Rule Set,?M??^?^?^?^?M??^?^?^?^?^ % 1.136% BUTTON HEAD SOCKET SCREW 6-32 X 1/2" HEX HEAD CAP SCREW 1/2-13 X 5 GRADE 5

28 ata Preparation Data Re-Engineering Step 2) Standardization Reference DB,, Reference DB

29 ata Preparation Data Re-Engineering Step 2) Standardization ID Description ======================================================= = A0059 GASKET,UPPER STEAM CYL,1929;; A0060 GASKET,DOOR,MG018;CASE;REACH IN A0061 MOTOR,BLOWER,3M726B;1/50 HP;SHADE POLE A0062 MOTOR,DRIVE,GOLD MEDAL;82085;POPCORN A0063 MOTOR,MAIN,NSK;110VOLTS; A TEETH,2-1/4" BORE,TAPERLOCK,50 PITCH,3-9/16", A0065 MOTOR,MAIN:20029;110VOLTS; A TEETH,1610 BORE,50 PITCH,MARTIN, ID Name # Val1 Val2 Val3 Diameter Threads... ======================================================= = A GASKET UPPER STEAM CYL A0060 GASKET DOOR MG018 A0061 SHADE MOTOR BLOWER A0062 MEDAL MOTOR DRIVE GOLD A0063 NSK 110 MOTOR MAIN VOLTS A0064 MARTIN 16 TEETH 2-1/4 IN BORE... A MOTOR MAIN 110VOLTS A0066 MARTIN 19 TEETH 1610 BORE 50 PITCH... Reference DB Description,

30 ata Preparation Data Re-Engineering Step 3) Matching Matching 3 1. Blocking : 1 2. Scoring : 3. Cutoff : Matching Blocking ( ID) Blocking ID Score MP DA DA X 700X 02 76X 700X 02 76X 700X XXX XXX XXX Cutoff (20 ) 37 DA X 700X XXX Scoring

31 ata Preparation Data Re-Engineering Step 4) Survivorship Matching 1/1/03 10/10/02 6/3/99 3 XG SOURCE 10/10/02 RECENCY 1/1/03 FREQUENCY Best-of-breed (Consolidated view) LENGTH 3 XG

32 Rule Data Data Rule Data Data Data Data Data

34 (P ) MES Data Quality Assessment PI Problem Solution Result ProfileStage AuditStage 10

35 (P ) As-is Source1 SAM File SAM File SAM File SAM Source2 LOGIC & Eye Check DB SAM File SAM File SAM File SAM Eye Checking Source

36 (P ) To-be Source1 SAM File SAM File SAM SAM AuditStage Source2 ProfileStage DB SAM File SAM File SAM SAM

37 (P ) (ProfileStage) Column Analysis Table Analysis Primary Key Analysis Cross-Table Analysis Relationship Analysis Normalization Analysis (AuditStage) Domain Analysis Completeness & Valid Assessment Structural Integrity Assessment

38 (Telstra ADBOR) Telstra Address DBOR Australia s largest Telco (one of the world s top 20) creates unique address verification system Problem Solution Result 170 Front-end / 100 Back-end Billing QualityStage Standardization, Matching Telstra Address Database of Record (ADBoR) Real Time API Telstra

39 (Telstra ADBOR) 1. OS/390 QualityStage RealTime Server 2., HTTP MQSerie s Data Services Search (C++) 5. DB Window NT DB

API QualityStage Real Time Server API Indices Target DB Presentation Layer

40 QualityStage RealTime Architecture QualityStage Application APIs Standardization, Verification, Matching, Survivorship etc. API QualityStage Real Time Server API Indices Target DB Presentation Layer SUPPORTED API s: Java, C, C++, COM, Cobol Server Layer SERVER PLATFORMS: UNIX, OS/390, NT Application Server Target Layer Database Server

41 2005 IBM Corporation

Agenda Stage Stage 7.5

Agenda Stage Stage 7.5 Ascential World Korea 2004 11 3 004 Ascential Software Corporation. All rights reserved. Ascential is a trademark of Ascential Software Corporation or its affiliates and may be registered in the United