Big Data - A Real Life Revolution

#tduniv

Big Data A Real Life Revolution

April 2014, Teradata Universe

Manuel Sevilla Global CTO

Business Information Management

@msevillatweets

2

Big Data & Analytics

Copyright © 2014 Capgemini. All rights reserved.

What is Big Data?

76 million

smart meters

in 2009…

200m by

2014

2+ billion

people on the

Web by end

2011

100s of

millions of

GPS enabled

devices sold

annually

4.6 billion

camera

phones world

wide

30 billion

RFID tags

today (1.3B in

2005)

12+ TBs

of tweet data

every day

25+ TBs

of log data

every day

Many PBs

of data every

day

80% Of world‟s data

is unstructured The 3Vs: Volume, Variety & Velocity

3



The real 3V explanation

Volume: BIG DATA

• More and more devices and

• Each device generates more and more data

Variety: ALL DATA

• Non structured and structured data together

• This is about an interconnected world with many external

partners working with no or low-modeled data

Velocity: FAST DATA

• This is about information and insights value

• Data and insight value decreases every minute!

1 0

0 1 0 1 0

0 1

0 1

1 1 0

1 0 1 1

0 0 1 1 1 0 1

0 1 0 0 1 1 0 1

1 0 1 0 1 1 0 0

1 0 1 1

4



In-memory has changed the game

An in-memory appliance

40 x86 cores, 1TB of RAM

For only 100 K EUR !

Moving a standard application from in-disk to

in-memory (without redesign) means a performance

increase between 100 and 1000 times faster

1 to 100 ratio: 2 minutes become 1 second

1 to 1000 means that

48 hours process may run in 3 minutes !

5



5760 cores for $3000 !!!

With two GK110 chips,

TITAN Z is powered by

a total of 5,760

processing cores, or

2,880 cores per GPU

Analytical

power is not

a limit

anymore !

6



From Predictive Asset Management…

Dust

from

dirty

wagons

Dynamic forces from trains

rubbing the ballast particles

together

Ballast choked

with fines – poor

drainage

The stiffness of

the underlying

formation

Voids under track

Embankments can degrade

through drying out and

shrinkage esp. in SE

Environment

dead vegetation

Strength of

underlying earthwork

Wet bed

7



to Predictive Human Healthcare

8



BI Appliances Hadoop

Expensive dedicated HW

Built for performance

Designed for high volumes (e.g. 10s of TB)

High availability (limited due to price pressure)

Initially developed using Relational Data bases

Very mature solutions (skills, SW, HW, administration)

Designed for modelled and structured data

Business As Usual ways to design, build and deliver

Teradata, Exadata, Netezza, HANA...

Commodity PCs

Built for extreme scalability (Batch oriented)

Designed for extreme volumes (10s of PB and more)

Very high availability

Initially developed for web ranking

Not as mature

Hadoop = Data is distributed over many machines

MapReduce = Computing is distributed and executed

where data is (grid solution)

9



Treatment Key

Object

SQL Database Search In

memory

Hadoop is first of all an extremely sustainable solution

Hadoop Distributed File System

Impala

Spark

HBase Map

Reduce

HAWQ SolR

SAS

HPA

HIVE, PIG, JAQL,

R support, ETL…

H

Catalog Stinger

Connectors Flume Storm Sqoop

http://www.oracle.com/index.html

10



IT World was split in 2 categories

Historical

Data

Streaming

Data

(Events)

GB

TB

PB

GB/s

MB/s

KB/s

Day Hour Min Sec SubSec

Data Warehouses

OLTP

Databases

Volume

Events Response Time

11



Big Data is about new Solutions, new opportunities

Historical

Data

Streaming

Data

(Events)

GB

TB

PB

GB/s

MB/s

KB/s

Day Hour Min Sec SubSec

Volume

Events Response Time

Data Warehouses

OLTP

Databases

Event Processing

Tools

Hadoop

In-memory

Databases

12



What is really innovative behind the Big Data buzz word ?

Hadoop & NoSQL : Low cost ways to store, manage and analyze

massive volumes of data

Cloud : Ability to rent, reduce Infra cost, reduce time to market

Event Processing tools : Ability to analyze and detect trends in real

time streaming events (monitoring, Next Best Action, Fraud...)

In-memory technologies : A new way to guarantee response time

even for very complex calculations

Explosion of Analytics usage : R is open source, High performance

statistics make them usable by every needed process

Multiple external data sources : Easier to tackle new data sources eg.

social media, traffic, GPS sensors, open data.

13



Multiple internal views – consistently compromised

Co

rpo

rate

Ad-h

oc

LO

B

ma

na

ge

me

nt

Op

era

tio

ns

Market

Op

era

tio

ns

LOB mart Spreadsheets

Line of business

Transactional systems

CRM ERP PLM

EDW

Fit

Detail

Freshness

Fidelity

Corporate

ODS

Web

14



The Business Data Lake Solution

HD

FS

Load everything

Keep the history

Business driven Country

Sales

Asian Marketing

campaign

Operations

data mart

Distill

Str

uc

ture

Transactional systems

CRM PLM ERP Meters Grid Web Social Media Market Supplier

15



You only need to be a few %

points better than your

competitors

16



The New Analytics Organization

A Proof of Concept (PoC) is usually delivered in months

Fail fast methodology!

This has to be considered as Research & Development : discovering new

patterns, new ways to reduce cost, improve customer experience, work on

Predictive Asset Maintenance, Fraud Detection

Massive use of Cloud

to reduce Time to Market

CAPEX vs OPEX , as a service mode

Very close to business and IT, often a Mixed team (internal + suppliers)

Dedicated to deliver new insights fast

Full access to many company data sources

Deep knowledge of IT data sources and modeling

Deliver temporary solutions, planed to be used as prototyping for

industrialization

Fast Analysts Teams

PoC Analysts teams

17



Data monetization : new revenue?

18



Data Privacy: 3 levels

May be federal, country or state level

May concern usage, storage, Information Lifecycle Management

May include legal requisitions (as with telco companies)

Level 1 : Law & Regulation

This is the way you want and you do not want to use all this information!

May be defined per BU, country…

May include partners / suppliers

Level 2 : Company level

Every customers wants be able to accept or reject any information usage

This may be linked to a contract, a checkbox, a call, a spam declaration…

This is not business as usual

You need to give to your customers the ability to define their own data access and

usage wishes (authorizations) !

Level 3 : Personal level

19



Our Big Data Service Center Framework

De

live

ry m

anag

em

en

t

Technical coordination

Service Center Governance

and PMO

Functional Coordination

Lean / Continuous

Improvement / Best

Practices

Functional & Technical

expertise

Factories governed by people, process and technology

POC Design Build RUN

Transformation – Big Data Business Cases

New Big Data development

Data usage (API, Analytics…)

Adding new data sources

Demand management Resource management

Quality and performance management

(Progress and quality)

Configuration and

release management

Infrastructure

architecture

Sandbox Link with legacy

Datacenter (countries)

Privacy management

Business Analysis

Datalab

Industrialization GO Live

Innovation

Big Data Governance

Design authority (Big Data architecture/roadmap)

BI vs Big Data governance

Change management

Prioritization and sponsorship

Transversal Delivery Strategy & Innovation

Infrastructure

Data Scientists

Analytics

Privacy & ILM

VM Package

Network Management

Architecture

Product / Solution choice

Delivery Model selection

Information Lifecycle

Management

Support and Administration

Analytics as a

Service

20



SMAC is dead. Long live SMAC!

SMAC New SMAC

Social Security

Mobility Movement

Analytics Artificial Intelligence

Cloud Creativity

21



Capgemini Global BIM Service Line Driving innovation with its technology partners

Capgemini „s global reach with operations in 44

countries and a focus on BIM with over 9600 BIM

practitioners.

A uniquely integrated approach to Information

Strategy based around the Capgemini “Intelligence

Enterprise”.

Deep Industry sector knowledge supported by Sector

Specific BIM offerings.

Capgemini‟s best-in-class Rightshore® capability for

BIM for development and management of BIM –

4000 BIM experts in India CoE.

A unmatched (and vendor independent) depth of

technology experience. Capgemini works with all the

major BI software vendors to deliver solutions

appropriate to the customer‟s needs.

850+ M EUR revenue in 2013

Europe:

South Africa

Argentina

Brazil

Mexico

United States

Canada

Saudi Arabia India

Australia

China

Morocco

Austria

Finalnd

France

Italy

Germany

Norway

Netherlands

Poland

Spain

Sweden

Switzerland

UK

http://www.oracle.com/index.html

22



Capgemini Group focuses on Big Data

Capgemini Group focuses on Big Data up to top level executive with his CEO, Paul Hermelin,

to support French Industry Ministry on Big Data project (34 French‟ Industrial projects)

www.redressement-productif.gouv.fr/files/la-nouvelle-france-industrielle.pdf#page=27

Paul Hermelin, Capgemini Group CEO, on live in

BFM Business® on 11/10/13 presenting Big Data

project for French Industry government .

http://pro.01net.com/editorial/605400/paul-hermelin-capgemini-l-etat-peut-etre-prescripteur-dans-le-big-data/

http://www.redressement-productif.gouv.fr/files/la-nouvelle-france-industrielle.pdf









23



Contact

Manuel Sevilla Global CTO

Business Information Management

@msevillatweets

#tduniv

The information contained in this presentation is proprietary.

Copyright ©2014 Capgemini. All rights reserved.

Rightshore® is a trademark belonging to Capgemini.

www.capgemini.com/bim

About Capgemini

With more than 130,000 people in over 40 countries, Capgemini

is one of the world's foremost providers of consulting, technology

and outsourcing services. The Group reported 2013 global

revenues of EUR 10.1 billion.

Together with its clients, Capgemini creates and delivers

business and technology solutions that fit their needs and drive

the results they want. A deeply multicultural organization,

Capgemini has developed its own way of working, the

Collaborative Business Experience™, and draws on Rightshore®,

its worldwide delivery model.

http://www.facebook.com/Capgemini

http://www.linkedin.com/company/capgemini

http://www.twitter.com/capgemini

http://www.youtube.com/capgemini

http://www.slideshare.net/capgemini

Big Data - A Real Life Revolution

Technology

open data

data variety

big data analytics copyright

fast data

worlds data

pbs of data

tduniv big data

structured data business