Date post: | 23-Jan-2015 |
Category: |
Documents |
Upload: | vicky-gillies |
View: | 415 times |
Download: | 0 times |
©2012 IBM Corporation
Understanding the real impact when
#Big Data meets #CloudReal world experiences, observations and findings
@LauraColvineLaura Colvine
Cloud Strategy Leader
IBM United Kingdom and Ireland
©2012 IBM Corporation|
The ‘art of the possible.’
©2012 IBM Corporation|
Big Data? - hype or reality?
©2012 IBM Corporation|
Big data characteristics in today’s digitised marketplace
4
Characteristics of big data
©2012 IBM Corporation|
The need for Cloud based Big Data and insight is
accelerating
60% potential
increase in retail operating
margins with big dataSource: McKinsey Global Institute: Big data, The Next Frontier for innovation, competition and
productivity, May 2011
of Global 1000
companies will store customer-
sensitive data in cloud by year-
end 2016.
50%
of Fortune 500
organisations will be unable to
exploit big data for competitive
advantage through 2015
85%
Source: Gartner: predictions for 2012
Source: Gartner: predictions for 2012
©2012 IBM Corporation|
‘Three out of four organisations have big data activities
underway’…. and patterns of adoption are forming
6
Total respondents n = 1061Totals do not equal 100% due to rounding
Big data activities
Respondents were asked to describe the state of
big data activities within their organisation.
Converging Data Architectures� Rebalancing data architecture portfolio,
blending compute and storage requirements
Context-based services� Where you are and what you are doing will
drive the next wave of digital services.
Consumable data services� The ability to share data will make it more
valuable--but only if it is managed differently.
Data for Insight and Impact � Visualisation and Discovery: Discover,
understand, search, and navigate federated
sources of big data while leaving that data in
place.
©2012 IBM Corporation|
Key Finding 1: Customer Outcomes and Optimisation are
driving big data initiatives across industry groups.
7
Consumer Goods Financial ServicesHealthcare /
Life Sciences
Manufacturing Public Sector Telecommunications
Customer-
centric
outcomes
Operational
optimisation
Risk / financial
management
New business
model
Employee
collaboration
©2012 IBM Corporation|
Convergence of physical, human and business process
data for better outcomes
Physical
Domain Operational
Domain
Human/ People
Domain
Consumers
Enterprises
Serving Aspirations of
Intergenerational
Consumer
©2012 IBM Corporation|
How are forward thinking organisations
using big data and cloud?
©2012 IBM Corporation|
Organisations are Putting Big Data and Big Insights
to work
Creating scalable, efficient,
and trusted information,
systems
Optimising complex
decision making, spot
trends and anomalies,
predict outcomes
Using resilient architectures
either on premise or in the
cloud.
©2012 IBM Corporation|
To Unlock the potential organisations master three
competencies to drive sustainable advantage
Align Anticipate Act
with confidence to optimise
service outcomesee, predict and shape
business outcomes
Organise, collaborate
and connect the people,
data and processes
©2012 IBM Corporation|
Align
Connecting Healthcare in the Cloud
Collaborating & Connecting People,
Organisations, Data and Process
Stadium in the CloudsEcosystems & Multi-Agency
©2012 IBM Corporation|
AnticipateUsing Data and Cloud to Provide
Actionable Insight
Reduction in wind forecast response time –from weeks to hours.
Vestas
97%Reduction in wind forecast response time –from weeks to hours.
Vestas
97%
717171
Growing from
2.5 PB to 6 PB
of data
©2012 IBM Corporation|
1200%Rizzoli Orthopaedic
30%Reduction in surgery-related hospitalisations
Identify Genetic Patterns: Enterprise Content Management
Data Warehousing, Application
Infrastructure, Cloud and IT
Optimisation
50%savings in data management costs .
Banco de Crédito del Peru
30% improvement in
transaction processing
efficiency
Increase in speed of collecting traffic data.
Bucheon City, South Korea
Act on Insights to improve service
outcome and customer satisfactionAct
©2012 IBM Corporation|
“Are we there yet?”....
©2012 IBM Corporation|
Convergance will increase as cloud stretches above IT
commoditisation into business optimisation
Next Generation Cloud
Cloud-Scale Data Challenges
Easy to Use Tools for Big Insights
Cloud based Social & Collaboration tools
Server and Storage Optimisation
Cloud Workload Analysis
Data Center Lifecycle Cost Analysis Tool
Security Analytic services
IBM Big Data/ Cloud Overview
Social
BusinessSustainabilityBig Data &
Analytics
Cloud
Computing SecuritySmarter
ComputingSmarter
Commerce
©2012 IBM Corporation|
There are Snakes and Ladders in the Big Data and
Cloud Discussion
Security
Compliance
Complexity
Workload Optimisation
Legacy & Transition
Skills & Culture
Cost
Business Innovation
Simplicity & Speed
Scalability
Collaboration
Customer Experience /Outcome
©2012 IBM Corporation|
Key Finding 2: Big data is dependent upon a scalable
and extensible foundation
18
• Multitenant Data Platforms
• Solid information
foundation
• Scalable and extensible
• Data in the Cloud
• Platforms for Data Analysis
• Platforms for Update
intensive workloads
• Data Platforms for Large
Applications
• Data Mash Ups
• Open Research Challenges
Big data infrastructure
Respondents with
active big data efforts
were asked which
platform components
were either currently
in pilot or installed
within their
organization.
©2012 IBM Corporation|
Engaging the Unengaged, Reducing risk and driving
revenue
S ecurity
Intelligence
Proficient
Proactive
Automated
Manual
Reactive
Proficient
Basic
Optim
isedMulti-Agency or
Ecosystems… Real-time,
Responsive, Aware
Single Issue or
Single Business
Function at this
level
………Structured
Data, or Unstructured
text data
Enterprise
Embedded & Edge
of Enterprise……Predictive Analytics
©2012 IBM Corporation|
From Siloed to Connected bridging the line of business
and IT perspectives
Example Enterprise Life Cycle
Strategy/Policy
Initia
tion
Ph
ase
Plan
Definition
Design
Re
aliz
atio
n
Ph
ase
Contract Definition
Realization/Build/Warranty
Operation
Op
era
tion
al
Ph
ase
Maintenance/Modifications
Disposal
Information Silos Typical Enterprise Functional Silos
Efficiency loss/cost estimated per li fe cycle phase/step due loss or lack of information between phases/steps
silo
silo
silo silo silo silo silosilo
silo silo silo silosilo
silo silo silosilosilosilo
silo
silo silo silo silosilo
silo silo silosilosilo
Efficiency loss/cost
estimated due loss or lack
of real-time information
integration between PA
(process automation) & OA
(office automation)
Efficiency loss/cost
estimated due loss or lack
of real-time information
integration between the
different enterprise silos
BI /
Reporting
BI /
Reporting
Exploration /
Visualization
Functional
App
Industry
App
Predictive
Analytics
Content
Analytics
Analytic Applications
BI /
Reporting
Exploration /
Visualization
Functional
App
Industry
App
Predictive
Analytics
Content
Analytics
Analytic Applications
Big Data Platform
Systems
Management
Systems
Management
Application
Development
Application
Development
Visualization
& Discovery
Visualization
& Discovery
Analytics AcceleratorsAnalytics Accelerators
Information Integration & GovernanceInformation Integration & Governance
Hadoop
System
Hadoop
System
Stream
Computing
Stream
Computing
Data
Warehouse
Data
Warehouse
©2012 IBM Corporation|
Deployment Complexity combines with the conflicting
needs of multiple stakeholders, each with specific
requirements
Data Acquisition
Streaming data
Text data
Multi-dimensional
Time series
Geo spatial
Relational
Data mining & statistics
Optimization & simulation
Fuzzy
matching
Network
algorithms
Composition and
PackagingCore Analytics
Filtering and
Extraction Validation
Social network
Video & image
Semantic
analysis
Business Rules Engine
Data Evaluation and FusionAlgorithm Composition and Invention
Testing and Execution Optimization
✔Deployment
New algorithms
Solutions
Data & Analytic
Services
Workload Optimization
Data & Analytic Runtimes
Data Models
Information Sources
©2012 IBM Corporation|
Workloads have specific characteristic that impact
Scalability, Optimisation and Resiliency design.
Late
ncy/B
iz D
ecis
ion T
hro
ug
hp
ut
D e gre e of A naly t ic C omple xit y
Hig
h/
Slo
wLo
w/
Ve
ry F
ast
Simple
(Alerts)Complex
(Simulation and Optimization)
Data
Co
mp
lexit
y
La
rge
nu
mb
er
of
da
ta f
ee
ds,
An
d/
or
lots
of
da
ta,
an
d/
or
un
str
uct u
red
da
ta
Lo
w
nu
mb
er
of
da
ta f
ee
ds,
An
d/
or
few
er
da
ta,
an
d/
or
Str
uctu
red
da
ta
Middle Road
(Forecasting)
Clickstream
analysis
Real-time game
monitoring
Cross-sales
Health
monitoring
Early w arning system
for energy trading
Health records
screening
Fraud
detection &
prevention
Capital market
surveillance
Card fraud
detection &
prevention
Lease
management
system
Retail inventory
optimization
Battlespace
command &
control
Call center monitoring
(cross sale)
Salesforce
enablement
Baggage
handling
Asset tracking
Telco QoS & SLA
monitoring
Trade
desk
monitoring
Liquidity
management
system
Manufacturing
process control
Shop floor
monitoring
Online hotel
booking
Call center monitoring
(quality)
Sensor based
water mgt
Intelligent Traffic
Systems
Industrial process
control
Telecom
billing
Real-time
Inventory
Optimization
Automated trading
Risk analytics
platform Seismic
Processing
Reservoir
Modeling
Astrophysical data
mining
Geospatial
tracking
Climate
Prediction
CAD/CAE
EDA
Weather Modeling
Telecom
netw ork
security (DPI)
Nuclear Energy
Simulation
Risk management in
energy trading
Massive Social
Media Analysis
Deep Q&A
©2012 IBM Corporation|
Cloud Systems of the future will be more data centric,
composable and scalable… but different data or
analytics workloads demand different system
characteristics
... there is NOT a 'one size fits all'
Cores SCM
StorageNetwork
Cores SCM
StorageNetwork
Cores SCM
StorageNetwork
Cores SCM
StorageNetwork
+ +
Predictive Analytics
Modeling, Simulation
Text Analytics
Hadoop Workloads
Optimisation
Sensitivity Analysis Future System
� Balanced, reliable, power efficient systems, with integrated software that scales seamlessly
� Integrated analytics, modeling and simulation capabilities to address generation, management and analysis
of Big Data for Business Advantage
General PurposeIntegrated Network
Integrated ProcessingIntegrated Storage
©2012 IBM Corporation|
Determining your highly valuable data from commodity
data will affect how you span on-premise and off premise
data workloads
Private cloud
forHigh Value Data
Sets that are Enterprise Unique
Public cloud
forData mashups,
Social Data,
Hosted cloud
forCommodity Data
sets with High Volume, High
Capacity
Beware the
economics of
data in cloud
In-House
forFor Organisations
with deep Big Data skills
©2012 IBM Corporation|
“Netflix, the movie-lender awarded $1m in 2009 to a team that improved the accuracy of its recommendation algorithm”
Source: DJ Patil building the data and analytics groups at
Facebook and LinkedIn.
The Emergence of the Data Scientist
©2012 IBM Corporation|
In Conclusion What have we learnt from our Big Data
cloud journey?
Plan an Information Agendato align with your strategy and
priorities
AlignYour Informationto govern the creation and
use of an integrated set of
accurate and relevant
information
Apply Outcome Analyticsto measure, anticipate and
shape business outcomes
Winners in the era of cloud and big data will be those who collaborate to unlock
data assets to drive innovation, make real-time decisions, and gain actionable
insights to be more competitive.
©2012 IBM Corporation|
Thank You