Data Strategy and
Architecture
The modern Your Data
data estate
Org Data CRM Graph Image Social IoT
leverages the Hybrid Ecosystem
best of on-
premise and
cloud On-premises Cloud
Private cloud
Management, Security,
and Insights anywhere
Operational databases Data Warehouses Data Lakes
Key Benefits
Reason over data, anywhere Flexibility of choice Security and Performance
challenges
Continuum to unlock digital innovation… Modernization
Digital Transformation
Data migration
Data modernization
Data Estate
Advanced analytics
Data intelligence
(AI/ML)
AND Convergence
Application Dev
Intelligent apps
(cloud native)
Smart apps (infused with pre-built AI)
App modernization
App migration
Foundational Innovation
Strategic value & capabilities
Data
Business
Use Cases
Operating
Model
Executive Technical
Strategy Capabilities
Develop an executive strategy based on the 3 guiding
principles that will enable a modern Data Estate
Modern Data Estate
Preparation Agility Resilience
Preparation
“Reduce Cost” Reduce costs though Increase your agility Be more resilient
Preparation through trusted insights to sudden change
People and Processes
Governance Architecture Data Lifecyle
“manage the system” “manage the container” “manage the content”
Executive • Charter • Data lake provisioning • Ingestion for compute,
• Tenets • Master Data Management • Handshaking
Strategy
Agility • Standards • Metadata Management • Control file
“Increase agility
Resilience • Data Quality • Common Data Model • Discovery (Data Catalog)
through trusted “Be more resilient • Data Access Management
• Security • Lineage
insights” to sudden change”
• Privacy • Data Contract
• Ethics • Linkage (Merge Service)
• Classification
• Retention
data strategy
Data strategy Data Strategy
Value unlock Business Applications and Reporting
Technology
enablers
Data Integration Data Processing & ML Data Access
Data Lakehouse
Foundations
Data Management and Governance
reference data architecture
Systems of
engagement
Data access
Data lakehouse
Data storage,
processing and
analytics
Data
integration
Data
management
System of
records
Current requirements Future requirements
“north star”
Description Example
Data architecture as an important organisational Reusable data pipeline to transform data from
asset that can be lifted and shifted that enable Azure Data Lake Service (ADLS) can be used other
several use cases use cases requiring similar patterns
Data that is ingested, stored and curated in the Leverage the rostering and planning data for both
data lake and database that can be utilised by use reporting and roster optimisation
cases other than that for which it was ingested
for
We use the reference data architecture and requirements to
map the components in the solution architecture
Key Requirements
Raw data PII columns will be hashed, 1. Raw data including data including Personally
Identifiable Information (PII) will stay in the source
and unnecessary columns will be
systems.
2. Azure Data Factory removes PII and perform pre-
removed prior to load in the data lake aggregation, if necessary, to de-identify the
information in batch e.g., daily, monthly.
3. Streaming services removes PII and perform pre-
aggregation, if necessary, to de-identify the
information in rea-time.
4. The de-identified data is stored in the data lake and
data warehouse in the cloud.
5. Databricks, Azure ML and cognitive services use de-
identified data to perform big data analysis and
machine learning.
6. Azure Purview is used to catalog and govern data
available on-premise and in the cloud.
7. Data can be access via API or other data connectors.
8. The Power BI and/or analysis services is used to
visualise the data into reports and dashboard
9. Business applications used data access layer to source
insights and data
ALM use case (phase 1) involves portfolio data for retail, EBP
and non-EBP extracted in batch
Thank you.
