Slide 1
Slide 1 text
Data Fabric
↓
The foundation
for a data-driven
enterprise
↓
From Data Fabric to
Data Observability
Slide 2
Slide 2 text
The rise of the
data-driven
enterprise
2
Data Fabric / © 2023 IBM Corporation
Slide 3
Slide 3 text
To be competitive,
you need to use data
as your differentiator!
Leverage your data to
create differentiated
experiences for
customers and
employees
3
Sales
Companies that are using
data-driven B2B sales-growth
engines report above-market
growth and EBITDA increases
in the range of 15 to 25 percent1
Marketing
Discover fast changing
customer behavior for
17% increase in
purchase
consideration.2
Operations
By 2025, nearly all
employees naturally
and regularly leverage
data to support their
work. Rather than
defaulting to solving
problems by developing
lengthy roadmaps.3
Finance
Set reliable forecasts in
volatile conditions for
90%+
improvement in time to
decision.4
HR
Attract and retain top
talent by engaging 600+
applicants a month.5
IT
Innovate and modernize
apps and infrastructure
for 25% more time to
focus on strategic
projects.6
Data Fabric / © 2023 IBM Corporation
Slide 4
Slide 4 text
21% 8%
However, evolving
to become data-driven
is challenging
4
use data-driven
decisions to realize
business value
“all the time”
have data and
advanced analytics
fully embedded into
company processes
and modernized
in the cloud
Source: Data-Driven Business Transformation,
Corinium 2022
Data Fabric / © 2023 IBM Corporation
Slide 5
Slide 5 text
As a result, enterprises
are missing out on
data-driven opportunities
52%
5
Data Fabric / © 2023 IBM Corporation
Revenue loss
or cost over-runs
43%
Inconsistency
in decisions
43%
Slow time to market
41%
Poor customer
experience
Slide 6
Slide 6 text
Because adoption of
hybrid and multi-cloud
environments requires
a new approach
6
– Attempt to centralize
everything fizzles
– Cloud and cloud
data warehouses
add complexity
– Higher compliance, security,
governance risks
– Increased focus
on regulatory requirements
– Multiple environments still
result in data silos
– Challenges related
to scale of:
• Discovering everything
you have
• Processing all your data
in disparate locations
• Data engineering
Data Fabric / © 2023 IBM Corporation
Slide 7
Slide 7 text
That new approach
is Data Fabric
7
Data Fabric (n):
A data architecture with an integrated set of
technologies and services designed to
democratize data access across the enterprise at
scale.
Data Fabric / © 2023 IBM Corporation
Slide 8
Slide 8 text
1.
Connect
2.
Profile
3.
Classify and enrich
5.
Publish
Establish
connectivity to
physical data
sources.
Assess the quality
of data assets.
Classify data assets,
assign data policies
and rules, and
enrich with
semantics.
Publish
data products
and enable
subscription to data
products based on
data contracts.
4.
Prepare /
transform
Engineer data
assets into trusted
data products.
Lifecycle and governance Implement DataOps principles throughout the
lifecycle and enforce governance end-to-end.
Data products
Data assets
Data Fabric / © 2023 IBM Corporation
Data Fabric enables a “trusted factory”
approach for innovation
Slide 9
Slide 9 text
AI governance
9
IBM’s approach
for Data Fabric
and AI lifecycles
Data governance
Data integration Data science
and MLOps
Data Fabric / © 2023 IBM Corporation
Slide 10
Slide 10 text
Deliver readily consumable
and reliable data to your teams
anytime and anywhere
03 Data integration
Data Fabric / © 2023 IBM Corporation
Slide 11
Slide 11 text
What is data observability?
02 Data observability
Data Fabric / © 2023 IBM Corporation
“Data observability” is the blanket term for understanding the health and the state of data in your
system. Essentially, data observability covers an umbrella of activities and technologies that, when
combined, allow you to identify, troubleshoot, and resolve data issues in near real time.
By encompassing a basket of activities, observability is much more useful for engineers. Unlike the data
quality frameworks and tools that came out along with the concept of the data warehouse, it doesn’t
stop at describing the problem. It provides enough context to enable the engineer to resolve the
problem and start conversations to prevent that type of error from occurring again. The way to achieve
this is to pull best practices from DevOps and apply them to Data Operations.
All of that to say, data observability is the natural evolution of the data quality movement, and it’s
making DataOps as a practice possible. And to best define what data observability means, you need to
know where DataOps stands today and where it’s going.
For more information, please goto https://databand.ai/data-observability/
Slide 12
Slide 12 text
Data Fabric / © 2023 IBM Corporation
Slide 13
Slide 13 text
Data monitoring vs data observability
02 Data observability
Data Fabric / © 2023 IBM Corporation
Slide 14
Slide 14 text
Why is data observability becoming a hot topic?
02 Data observability
Data Fabric / © 2023 IBM Corporation
Multi-cloud organizations
The field is changing rapidly…
Digitization leads to more data integrations
Data is becoming client-facing by default
… leading to increased complexity
Reduced efficiency
Large dependency on experts
Higher business risk with data issues
Monitoring & act real-time
Data in motion & 24/7 delivery
Slide 15
Slide 15 text
02 Data observability
Data Fabric / © 2023 IBM Corporation
Big FMCG client
“
”
One of our largest challenges is maintaining and tracking the
data quality that is streaming through our platforms as it is
every increasing.
We want to make sure our data scientists can focus on the
work they love instead of monitoring and spending a lot of time
in debugging
Slide 16
Slide 16 text
What can data observability deliver?
02 Data observability
Data Fabric / © 2023 IBM Corporation
DETECT EARLIER RESOLVE FASTER DELIVER TRUST
Pinpoint unknown data incidents and reduce
mean time to detection (MTTD) from days to
minutes
Improve mean time to resolution (MTTR)
with incidents alerts and routing from weeks
to hours
Enhance reliability and data delivery SLAs,
provide visibility into pipeline quality
issues.
Slide 17
Slide 17 text
Building a central data observability center to
achieve optimal data solutions
02 Data observability
Data Fabric / © 2023 IBM Corporation
IBM Databand
Data Observability
Incident management
Anomaly detection
Custom alerting
Alert centralization
Data pipeline monitoring
Data quality monitoring
Impact analysis and lineage
Root cause analysis
Ingestion, orchestration,
transformation
Data warehouses
Data lakes / lake houses
Dashboard & reporting
Data Teams
Real time alerts and SLAs dashboards for rapid resolution.
Notifications
Data engineering
Data platform
Analytics
Automated metadata collection and error logging.
Slide 18
Slide 18 text
The Databand solution
02 Data observability
Data Fabric / © 2023 IBM Corporation
1. Collect
Automatically collect metadata.
From all key solutions in the
modern data stack.
4. Resolve
Automatically collect metadata.
Create smart communication
workflows to resolve data quality
issues and meet SLAs.
2. Profile
Build historical baseline.
Based on common data
pipeline behavior.
3. Alert
Alert on anomalies and rules.
Based on deviations or breaches.
Data Fabric / © 2023 IBM Corporation
Slide 19
Slide 19 text
How does a standard implementation flow look like?
02 Data observability
Data Fabric / © 2023 IBM Corporation
Design Translate Test Scale
1. Map current architecture
2. Determine adjustments
3. Identify major pain points
4. Decide on area for PoC
5. Translate into Tech & Functional doc
Slide 20
Slide 20 text
Data Fabric / © 2023 IBM Corporation
Take care of your data
observability and leverage the
gained time to expand your
business!
Slide 21
Slide 21 text
Any questions?
For more information you can find us at the following stands at the data expo
Implementation partner
Standnr. 5 Standnr. 136