Watch full webinar here: https://bit.ly/3xdSTIU Digital Transformation has changed IT the way information services are delivered. The pace of business engagement, the rise of Digital IT (formerly known as “Shadow IT), has also increased demands on IT, especially in the area of Data Management. Data Services exploits widely adopted interoperability standards, providing a strong framework for information exchange but also has enabled a growth of robust systems of engagement that can now exploit information that was normally locked away in some internal silo. Join us for our upcoming Middle East Webinar series episode, “Data Services and the Modern Data Ecosystem,” presented by Chief Evangelist MEA, Alexey Sidorov. Tune-in as we explore how a business can easily support and manage a Data Service ecosystem, providing a more flexible approach for information sharing supporting an ever diverse community of consumers. Watch on-demand this webinar to learn: - Why Data Services are a critical part of a modern data ecosystem - How IT teams can manage Data Services and the increasing demand by businesses - How Digital IT can benefit from Data Services and how this can support the need for rapid prototyping allowing businesses to experiment with data and fail fast where necessary. - How a good Data Virtualization platform can encourage a culture of Data amongst business consumers (internally and externally)

1. Data Services
in the Modern Data Ecosystem
Alexey Sidorov PhD
Data Management Director, Chief Evangelist

2. § Agenda
§ Modern Data Ecosystem
§ Practices & Patterns
§ Use Cases
§ Demo
§ Closing Remarks
§ Q & A

3. 3
Modern Data Ecosystem

4. 4
Reporting
Analytics
Data
Sources
Copy
Staging
Area
Copy
Core
DWH
Copy
Data
Marts
Duplicated Data
5-6 hours on average
to move data from
sources to data marts

5. 5
Businesses increasingly demand
capabilities that enable them to
achieve digital transformation
through faster integration, real-time
analytics, and better data discovery
process

6. 6
While IT keeps trying to get all data
to a single repository, that data has
grown across systems, transcending
data warehouses, data lakes, cloud,
and, more recently, to the edge.

7. § Tectonic Shift in Data Requirements
Deployment On-premise Cloud/Hybrid
Processing Batch Real-time
Access Point-to-point Decoupled
Data Models Rigid Flexible
Orchestration Manual Automated

8. Data Services
AI / ML
Reporting
Analytics
Real-time Alerts
Operational AI
Analytical Data
Operational Data
Streaming Data
>>>
§ Modern Data Ecosystem

10. 10
Instant access to everything
without needing to store anything

11. 11
Practices & Patterns

12. § The Relevance of APIs Ecosystem
§ The internet has created an interconnected
world
§ Similarly, different processes and
applications within a company also need to
communicate with each other
§ Web services are the building blocks of this
interconnected world
§ The concept of “an application exposing
functionality” has evolved into “the web
service is the application” (microservices)

13. § Definitions
§ API - an interface
§ Web Service - a remote API via the web
§ Data Service - a web service for data
§ Microservice - an architectural style

14. Data Services
AI / ML
Reporting
Analytics
Real-time Alerts
Operational AI
Data Virtualization
Data
Domains
Universal
Catalog
of
Data
Services
Centralized
Access
Control
Virtual
Datamarts
Raw
Data
Analytical Data
Operational Data
Streaming Data
>>>
§ Modern Data Ecosystem

15. § Common Scenarios for Data Services

16. 16
Data
Science
Machine
Learning
Artificial
Intelligence
Third Party
Applications
Predictive
Analytics
Business
Intelligence
Data Services
Virtual Data Domains
Granular Security
SQL
Self-Service
Data Catalog
Relational NoSQL Unstructured Docs Cloud Sensors IoT
Sources
Universal Access
API GraphQL OData
Party
Product
Customer Location Employee
§ Modern Virtual Data Architecture

17. § Patterns
§ Aggregation Offload processing from front-end to back-end. Exploits pushdown
capability. Also, can be used in conjunction with caching and/or
query acceleration.
§ Fusion New fields that contain data expanded from existing. Often done to
avoid storing or modifying the underlying system. Examples include
adding geospatial info or formulas sourced from multiple columns.
§ Blending Service that incorporates multiple elements together. May specify
rules for what/when to blend. Data exists in separate repositories
linked by unique identifier(s).
§ Filtering Often driven by security to limit (rows) or restrict (columns) based on
the type of requestor. When used for performance, becomes
Microservice Architecture Pattern.

18. § Patterns
§ Domains Enable microservices to be grouped together to express particular
(functional) domain areas. Often driven by ownership (centralized or
distributed) and control (e.g. official/curated). Leverage traversable
relationships.
§ Composite Abstract complexity of underlying services, including drivers for
security, transactionability, performance, and modeling.
§ Sharing Typically used to expose “data” to “application” tier or from “internal”
organization to “external.” Often used in association with an API
Management tool and use of Web-based IdP (SAML, OpenID, and
OAuth2) solutions.

19. § Capabilities for Data Services
§ Data models (tables, views, stored procedures) available
automatically as web services - zero coding required
§ Available in multiple formats: RESTful (XML, JSON), OData 4,
GeoJSON
§ Support for GraphQL: flexible new format for data services
§ Automatic documentation (OpenAPI) and integration with Data
Catalog
§ Authentication with modern protocols like OAuth 2.0
§ Authorization based on roles with, including column/row
restrictions and masking
§ Workload management: priorities, quotas (queries per hour),
restrictions by user/role/IP, etc.
§ Caching and query acceleration capabilities
§ Monitoring, access auditing

20. 20
Use Cases

21. 21
Perth DC
Brisbane DC
Santiago DC
Houston DC
US East Cloud
APAC Cloud

22. 22
Demo

23. 23
Closing Remarks

24. § Key Takeaways
§ Data Virtualization enables reduced time-to-market and
improved data asset utilization via APIs in modern data
ecosystems
§ Decoupling access and storage is a fundamental concept
with APIs and Data as a Service
§ Real-time is especially important when interacting with
business processes and analytic models
§ Microservice approaches like REST, OData and GraphQL
augment data use

25. 25
Gartner Data Virtualization Market Guide, Dec 2018
Through 2022, 60% of all organizations will
implement data virtualization as one key delivery
style in their data integration architecture”

26. 26
Kevin Kelly The Inevitable
Digital Transformation requires the
mindset shift …
Sharing data is more effective than
accumulating

27. Q & A

28. Thanks!
www.denodo.com info@denodo.com
© Copyright Denodo Technologies. All rights reserved
Unless otherwise specified, no part of this PDF file may be reproduced or utilized in any for or by any means, electronic or mechanical, including photocopying and microfilm,
without prior the written authorization from Denodo Technologies.
Alexey Sidorov PhD
asidorov@denodo.com
+971 56 625 3066
+7 915 185 5562