12/9/23, 10:01 PM Data Fabric 2.

0 For Connected Intelligence

Data Fabric 2.0 For Connected Intelligence

Topics

ROADMAP REPORT New Data Challenges Are D… Data Fabric 2.0 Offers New … Data Fabric 2.0 Goes Beyon…

Data Fabric 2.0 For Connected Intelligence

Forrester’s Reference Architecture For Data Fabric
February 16, 2023

NY Noel Yuhanna

with Aaron Katz, Michele Goetz, Karsten Monteverde, Kate Pierpont

Summary

Forrester’s data fabric vision has evolved over the years as enterprises have implemented various use
cases and vendors have extended products to deliver broader and innovative capabilities. In updating
our data fabric reference architecture to version 2.0, we reflect new technology enhancements
supporting native cloud, advanced data intelligence, dynamic services, data products, and global
transactions. Technology architecture professionals should look at Forrester’s data fabric 2.0 and use it to
guide their data and analytics strategy.

New Data Challenges Are Disrupting Enterprise Data And

Analytics Strategy
Business users need accurate, real-time insights to support business decisions like next best
experiences, churn identification, and media optimization. However, traditional data architecture
lacks the agility that today’s business environment requires to deliver fast, consistent, and trusted
data across silos (including applications). This aligns with why Forrester’s 2022 data shows that
data and analytics decision-makers cite data management and data security among their top
obstacles to achieving their data and analytics vision (see Figure 1).

Data fabric emerged to address this gap between traditional data architecture and emerging
business needs. We are updating our data fabric framework and reference architecture to version
2.0 to reflect the addition of data fabric capabilities that accommodate a range of new data sources
(e.g., sensors, social media, smart devices) and rapidly evolving business requirements, including:

Establishing consistency and trust for distributed and edge data. Distributed data stored
across on-premises and multiple clouds has led to data duplication and challenges around
consistency, integrity, and security. The lack of consistent and trusted data across data centers
and clouds is slowing down innovation and growth, impacting new insights, and degrading
customer experience. Moreover, data gravity is pushing data out to the edges, and making it
more challenging to leverage data from systems of engagement together with systems of
record.

About Forrester Reprints https://go.forrester.com/research/reprints

https://reprints2.forrester.com/#/assets/2/108/RES178910/report?ocid=eml_pg404321_gdc_comm_az 1/5
12/9/23, 10:01 PM Data Fabric 2.0 For Connected Intelligence

Protecting sensitive data to meet tougher regulatory and compliance mandates. Enterprises
Data Fabric 2.0 For Connected Intelligence
are under pressure to deal with data privacy regulations, such as GDPR, CCPA, and HIPAA,
requiring
Topics more advanced and granular data security and protection measures. Today, many
data platforms are highly vulnerable due to poor authentication practices, the use of default
New Data Challenges Are D… Data Fabric 2.0 Offers New … Data Fabric 2.0 Goes Beyon…
configurations, and weak data access control measures. As a result, enterprises are spending
more time and effort protecting sensitive data, which can slow down (or stop entirely) new
data sharing, collaboration, and analytics initiatives.

Supporting real-time decisions with real-time data. Data assets are spread across multiple data
centers and clouds, making it difficult to get real-time data to support new and emerging
business operations. Traditional data architectures were not designed to support real-time
data, especially in a distributed environment. Enterprises often cite support for real-time data
as a top data initiative, primarily when supporting microservices applications, real-time
insights, and customer experience applications.

Integrating the growing ecosystem of external and third-party data. The explosion of public
and external data, including property records, credit card data, government open-data
initiatives, and consumer-generated content on social media has created a new business
opportunity for organizations. Although this treasure trove of data can deliver deep new
insights into customer and market behavior, it requires a new and open means of integrating
those data with operational and other sources.

Figure 1

Data Management And Data Security Headline Main Challenges For Organizations

About Forrester Reprints https://go.forrester.com/research/reprints

https://reprints2.forrester.com/#/assets/2/108/RES178910/report?ocid=eml_pg404321_gdc_comm_az 2/5
12/9/23, 10:01 PM Data Fabric 2.0 For Connected Intelligence

Data Fabric 2.0 Offers New Capabilities To Accelerate And

Data Fabric 2.0 For Connected Intelligence

Simplify
Topics Deployments

New Data Challenges Are D… Data Fabric 2.0 Offers New … Data Fabric 2.0 Goes Beyon…
Data fabric is a modern architecture that automates the integration of any data in real time or near
real time from disparate sources, on-premises or in the cloud, into coherent data services that
support business transactions, analytics, predictive analytics, and other workloads and patterns. It
minimizes complexity and hides heterogeneity by embodying a cohesive model of data that reflects
business requirements rather than the details of underlying systems and sources. The fabric can
adapt to changing business policies and rules while delivering trusted data.

With early implementations, organizations relied on systems integrator (SIs) to help integrate,
orchestrate, transform, prepare, and secure data and manually integrate multiple products and
technologies to create a fabric-like architecture. Over the past decade, implementation patterns
became commercialized in a range of data fabric products from Denodo, Oracle, SAP, Talend, IBM,
and Informatica, which provided more integrated solutions. Since then, mainstream usage of data
fabric has grown, and product implementations have matured and become more innovative,
embracing emerging trends like graph engine, streaming, data intelligence, distributed in-memory,
and more robust integration with cloud-native.

Data fabric 2.0 offers several new capabilities, including (see Figure 2):

Built-in data intelligence that accelerates deployment. Although data intelligence has been in
the data fabric since its inception, data fabric 2.0 delivers more advanced capabilities to
automate data management. These capabilities include intelligence to discover sensitive data
automatically, recommend integration of known data elements, and transform data based on
specific data patterns. Forrester expects more data intelligence to evolve within the data fabric
to simplify deployments across all use cases.

Integration with master data management (MDM) to deliver trusted data. Although data fabric
is not MDM, it can be used along with MDM solutions to deliver a consistent model of
information to support business needs. The data fabric separates the technical concerns of
distributed data access and management from the business concern of managing consistent
master data. It also provides a place where the model defined by MDM can be embodied in
virtual form, ready for access through industry-standard interfaces for applications and
insights.

Graph engine that automates and connects data at scale. When you connect five or 10 data
sources, connecting them manually by writing code in Java, Python, or C# is doable. However,
connecting hundreds or thousands of data sources can become a nightmare. The graph is the
fastest way to connect data as opposed to manually connecting them, especially when
dealing with large or complex data sets. The graph uses nodes, edges, and properties to
discover connected data in an automated manner within a data fabric. Without graph
technology, it can be slower to connect the dots to support data integration and orchestration.

Real-time data quality and transformation to deliver consistent data. Real-time quality cleanses
data on the fly based on policies, configurations, or events by alerting, flagging, or removing
data that failed to meet the defined data requirements. Real-time transformation implements
data fabric policies to change or enhance real-time data based on events or context, such as
weather, stock market performance, customer behavior, or location.

Global transaction management. Although early data fabric deployments were mostly read-
only, the adoption of transactional fabrics is growing rapidly, especially to support modern
mobile and cloud apps. Transaction management in the data fabric orchestrates data between
the fabric and sourced data repositories, whether internal or external, to ensure data integrity
once events commit or roll back a unit of work.
About Forrester Reprints https://go.forrester.com/research/reprints

https://reprints2.forrester.com/#/assets/2/108/RES178910/report?ocid=eml_pg404321_gdc_comm_az 3/5
12/9/23, 10:01 PM Data Fabric 2.0 For Connected Intelligence

Prebuilt multiple domains to accelerate business use cases. Although data fabric has been
Data Fabric 2.0 For Connected Intelligence
supporting domains since its inception, data fabric 2.0 offers prebuilt multiple domains to
accelerate
Topics deployments. These include domains to support customers, products, suppliers,
employees, and partners. Organizations can choose between building each data fabric
New Data Challenges Are D… Data Fabric 2.0 Offers New … Data Fabric 2.0 Goes Beyon…
instance with a domain or having multiple domains within a single data fabric.

Additional support for data products. Data fabric has been enabling data products for years,
facilitating data products with integrated data and metadata based on semantics and policies,
which can be leveraged in BI, embeddable applications, or accessed using APIs. In data fabric
2.0, there is more support for nontechnical roles such as business analysts and business users
to build data products directly with data fabric, enabling the creator community.

Figure 2

Forrester’s Data Fabric 2.0 Reference Architecture

Data Fabric 2.0 Goes Beyond Customer 360

Enterprises don’t implement data fabric for just one use case; it is an enterprise architecture that
can support dozens of use cases, including customer 360, fraud detection, the internet of things
(IoT), edge, and real-time analytics. Although the use cases for data fabric continue to expand, we
find enterprises most commonly using data fabric 2.0 for:

Data mesh. A third of the data fabric vendors report that their new customer deployments
support a data mesh architecture. Data mesh is typically deployed using multiple technologies,
including data fabric. With data fabric 2.0, organizations are building domain-specific fabric,
building data products, and leveraging data fabric interconnect to connect to other fabrics
through federated governance.

Global operational insights. A new generation of business apps is serving the need for global
data, leveraging microservices, edge applications, and global operational workloads. For
example, a smart mobile banking app may keep track of customers’ activity across countries,

About Forrester Reprints https://go.forrester.com/research/reprints

https://reprints2.forrester.com/#/assets/2/108/RES178910/report?ocid=eml_pg404321_gdc_comm_az 4/5
12/9/23, 10:01 PM Data Fabric 2.0 For Connected Intelligence

which requires the app to leverage consistent, trusted, and real-time data. Data fabric 2.0
Data Fabric 2.0 For Connected Intelligence
delivers low-latency access to critical data across data centers, regions, and clouds to support
global
Topics operational workloads.

Data science (AI/ML).

New Data DataAre
Challenges fabric
D… 2.0 makes
Datadata
Fabricscientists more
2.0 Offers New … productive byFabric
Data helping them
2.0 Goes Beyon…

provision and prepare modeling data faster with trusted, integrated, and transformed
connected data in a single logical data platform. Although less than 7% of data fabric
deployments are used for data science workloads today, the adoption is likely to double in the
next two years.

Distributed transactions. Global transactional applications can leverage localized data with
distributed data fabric 2.0, delivering low-latency access to consistent critical data across
regions. For example, businesses might need to support hotel reservations, flight bookings,
expense reporting, or other applications across regions where the data and transactions must
be consistent and real time.

Data sharing/collaboration. Data sharing among business users can be challenging, especially
when distributed data requires granular security controls to meet compliance mandates and
data must be available 24/7. Data fabric 2.0 offers a platform to store, process, and access
common and integrated data based on domains and context to support visualization,
dashboarding, and SQL queries.

© 2023, Forrester Research, Inc. and/or its subsidiaries. All rights reserved.

About Forrester Reprints https://go.forrester.com/research/reprints

https://reprints2.forrester.com/#/assets/2/108/RES178910/report?ocid=eml_pg404321_gdc_comm_az 5/5