Enterprise Master Data Management (MDM)

Introduction
•Goal: Single place where all common master data in an organization is stored and managed. The data would be accurate, consistent, and maintained in a coherent and secure manner.
•Provides a consistent understanding and trust of master data entities
•Provides mechanisms for consistent use of master data across the organization
•Is designed to accommodate and manage change

Why MDM
•Cross-LOB Perspective (Investments, Loans, Deposits; sees data as critical to operations, not see value in sharing)
•Cross-Channel Perspective (dist. Channel- Partner, Internet, Branch; different solutions->account-centric->customer centric))
•Cross-Business Subdomain Perspective (Case history, Contact preference, Party; different scope)
•Cross-Application/Technology Perspective (packaged apps; variance in technoical platform
•Mergers and Acquisitions

MDM System
•Master Data Domains – Recognition of CDI and PIM. Three primary domains: party, product and account
•Methods of Use – Collaborative Authoring (users & systems to reach agreement), Operational (providing stateless services), Analytical (trusted data source, key function or analytics)
•System of Record (read-only) vs. System of Reference
•Consistency of Data – Absolute Consistency (consistent all the time), Convergent Consistency
•Implementation Styles – Consolidation Implementation (gold source); Registry Implementation (for read-only); Coexistence Implementation (master in many locations); Transactional Hub Implementation.
•Categorizing Data – Metadata, Reference Data, Master Data, Transaction Data, Historical Data

Collaborative MDM (New Product Introduction example)
•1. Receive Notification of New Item
•2. Create Draft Item
•3. Classify Item and Assign SKU
•4. Define Item Properties
•5. Define Marketing Properties
•6. Assign Item to Locations
•7. Define Finance Properties
•8. Approve Product Definition

Operational MDM (OLTP)- New Account Opening
•1. RECORD Arrangement Request
•2. ANALYZE Customer RElationship
•3. ANALYZE Arrangement Request
- APPLY Product Policy
- APPLY Credit Rating Scale
- FORECAST Arrangement Risk
- OFFER Arrangement

Implementation Style
•Consolidation
•Registry
•Coexistence
•Transactional Hub

Data Category
•Metadata
•Reference data
•Master Data
•Transactional Data
•Historical Data

Business Benefits of MDM
•Consistent Understanding and Trust of Master Data Entities – Accuracy, Completeness, Consistency, Timeliness, Relevance, Trust
•Consistent Use of Master Data Across the Organization – Cost Savings and Efficiencies, Regulatory Compliance
•Accommodate and Manage Change – Reducing Time to Market, Revenue Enhancement and Other New Opportunities, Ability to Rapidly Innovate, Product or Service Innovation, Process Innovation, Market Innovation, Supply Chain Innovation, Accommodating Mergers and Acquisitions, Introduction of New Requirements

An SOA Enabler (SOA Architecture)
•Layer 1 – Consumers
•Layer 2 - Business Process (Composition, Choreography, Business State Machine, Orchestration)
•Layer 3 – Services (atomic and composite)
•Layer 4 – Service Components
•Layer 5 – Application Services
•Layer 6 – Data Repositories & Information Services
•Layer 7 – Integration (Enterprise Service Bus)
•Layer 8 – Quality of Service (Security, Management, Monitoring)
•Layer 9 - Governance

Characteristics of SOA services (also for MDM)
•Service reuse
•Service granularity
•Service modularity and loose coupling
•Service composability
•Service componentization and encapsulation
•Compliance with standards (both common and industry-specific)
•Services identification and categorization
•Provisioning and delivery
•Monitoring and tracking

Information as a Service: Characteristics
•Definition – The structure and the semantics of the information needs to be well defined and commonly available
•Quality – integrity of the data needs to be ensured for retrieval and update
•Governance – Changes to the service and the underlying information need to be governed in a uniform and consistent manner

MDM Reference Architecture
•Conceptual Level
•Logical Level
•Physical Level

Architecture Pattern
•Process-Focused Application Integration (integration of applications)
•Information-Focused Application Integration (synchronize master data among MDM hub and underlying legacy systems)
•MDM Hub Patterns (style of MDM deployment)

MDM Ref – Key Functional and Technical Capabilities
•Master Data Lifecycle Mgmt Capability – from created to no longer required; group and define hierarchies; flexible mapping; define master data hierarchies, relationships, groupings; versioning; model multiple taxonomies; authoring; security; audit;
•Data Quality Mgmt Capability – analysis & profiling; standardization, data validation, data cleansing logic; Data reconciliation; data governance; measure the staleness of data
•Master Data Harmonization Capabilities – Integration (messaging, service invocation, batch, ETL, FTP); error-handling; support high-volume transaction
•Analysis and Insight Capabilities – discover insightful relationships; improve business decision; access structured and unstructured information; manage the state of a process; configure event management services

Conceptual Architecture
•Framework to manage and maintain master data
•Scalable, highly available, adaptive architecture
•Coordinate, manage the lifecycle of master data across the enterprise
•Accurate critical business information available as a service
•Cleanse data, improve the quality and consistency of master data
•Make master data active by detecting events and generating operations to manage master data
•enable the ability to implement solutions

MDM Solution – Key architecture building blocks
•Third-Party Data Service Provider
•Process Manager to choreograph
•Connectivity and Interoperability Layer
•MDM Services and Master Data Repository
•Information Integration Services
•Identity Analytics

MDM Solution – Architecture Principles
•Provide ability to decouple information from app & process
•Available as a strategic asset for enterprise
•Authoritative source for master data (manage integrity, control distribution in a standardized way)
•On an architectural framework and reusable services
•Based on industry-accepted open computing standards
•Provide flexibility to accommodate changes
•Highest regard for preserving the ownership of data, integrity and security of data
•Ability to incrementally implement a MDM Solution

MDM – Logical Architecture Components
•Interface Services
•Lifecycle Management Services
•Hierarchy and Relationship Mgmt Services
•Master Data Mgmt Event Mgmt Services
•Authoring Services
•Data Quality Mgmt Services
•Base Services
•Master Data Repository

MDM – Information Risk Analysis
•Identifying the information assets
•Assigning value to each asset
•Identifying each asset’s vulnerabilities and associated threats
•Calculating the risk for the identified assets
•Evaluating different countermeasures in terms of costs and reduction of risk they provide
•Recommending the appropriate countermeasures

MDM – Security and Privacy: Types of IT Risks
•Operational risks – failures in business process; denial of service attack;
•Regulatory and Compliance risks – meet business processes; adequately protecting sensitive data
•Reputational Risks -

MDM – Information Risk Management
•Risk Analysis for MDM
•Security Control Selection and Implementation

MDM – Identifying MDM Assets
•Sources of master data
•Master data itself
•Consumers of master data
•Other related assets

MDM – Security Consideration
•Policy
•Confidentiality
•Integrity
•Identity
•Authentication
•Authorization
•Audit
•User Registry
•Identity provisioning
•Identity token
•Identity mapping
•Identity Propagation
•Reverse Proxy

MDM – Security Considerations
•Identity Propagation, Mapping & Provisioning – Business (Trust Mgmt, Identity and Access (Authorization) Mgmt); Technical (Identity & Authentication Service, Policy Mgmt)
•Authorization – Business (manage identities, roles and groups); Technical (standards-based to handle specifying, distributing & enforcing authorization policies.
•Audit – Business (comply with policies & reports illustrating how well relative to policies; Technical (audit events, real-time and post-processing events reporting
•Data Protection – Business (describe business object level how master data should be protected); Technical (encryption, SSL, WS-Security)

Logical SOA Security Architecture
•Business Security Services tier
•Security Policy Mgmt tier
•IT Security Service tier

Security Enablers
•Cryptography
•Key Management
•Hardware key Storage
•Cryptographic Hardware
•Malware Protection
•Isolation
•Firewalls
•Intrusion Detection
•Intrusion Prevention
•Time
•Security Event and Incident Mgmt (SEIM)

Policy Management
•Policy abstraction level
•Policy management lifecycle
•Policy Domains

Identity propagation
•Security Token Service (STS)
•SAML token for security token format

Authentication Services
•WS-Trust Security Token Service (STS)

Authorization Services
•Service Consumer
•MDM SOA Services Layer
•MDM Services Implementation
•Master Data Repositories