Transcript Unleashing enterprise mobility

www.pwc.com/techforecast
Data and analytics trends
Discussion at the FEI CFIT Event
Dec. 13, 2012
Technology forecast series on data and analytics
Semantic Web
Modeling and
simulation
A lot more context and content than what’s in this deck
PwC CTI
Big Data
http://www.pwc.com/techforecast
New analytics
Trends we’ll be discussing
More clues in the data web
•
Semantic integration layers
•
Social and interest graph data
•
Crowdsourced data description
Mining clues in the data web
•
The boom in database heterogeneity and NoSQL/NewSQL
•
Hadoop as a preprocessing environment
•
Big Data in 2012 and embedded Hadoop
The three waves of analytics
•
Central
•
Self-service
•
Cloud
PwC CTI
http://www.pwc.com/techforecast
3
More clues in the data web
PwC
The boom in less structured, external data
PwC CTI
http://www.pwc.com/techforecast
Data feudalism versus the webby way
Data feudalism
Semantic integration
App centric
Content/data centric
Monolithic packages
Services, then URIs as APIs
Developer lords, manor system
Linked data cloud
Dumb data
Logic embedded in the web
Looking under the lamppost for
your keys because that’s where the
light is
Scaling an integration layer
PwC
Crowdsourcing semantic metadata + data
Linked Data: 2009 vs. 2011
2009
PwC CTI
2011
http://www.pwc.com/techforecast
How machines read across domains
Non-scalable
PwC CTI
Scalable
http://www.pwc.com/techforecast
Crowdsourcing structured content via Twitter
1
Magnum Photos have just six staff members to
tag 6,000 museum-quality photos a month being
added to their repository.
December 2012
PwC
2
Tagasauris helps Magnum crowdsource the
effort, slicing up the work into 23 microtasks.
Slide 9
Enabling discovery through data description
3
Tagasauris encourages people via Twitter to join
the tagging effort by scoring their work and
making their accomplishments visible to the
community.
December 2012
PwC
4
Magnum now makes it possible for search
engines and thus potential users to find more
than 1 million of their photos online.
Social and interest graph visibility—contextual
web emerging
1
3
2
More explicit, machine-readable
relationships between people, places
and things should help with relevance
and thus the overload problem.
PwC CTI
http://www.pwc.com/techforecast
11
How the
social/interest graph
can link silos
PwC CTI
http://www.pwc.com/techforecast
12
Devil’s advocate discussion #1
Why bother structuring web data?
Can’t brute force statistical analysis tell us everything we need to know?
We’ve heard of the semantic web, but have never seen any results. Will
it ever go mainstream?
PwC CTI
http://www.pwc.com/techforecast
Mining clues in the data web
PwC
http://www.pwc.com/techforecast
14
Tackling gray data
PwC CTI
http://www.pwc.com/techforecast
Slide 15
Hadoop as a preprocessing environment
PricewaterhouseCoopers, derived from Apache Software Foundation and Dion Hinchcliffe, 2010
PwC CTI
http://www.pwc.com/techforecast
Slide 16
Case study
Disney: treating different data differently,
without a lot of extra investment
“My software cost [for the D-cloud data cluster] is zero.
You still have the implementation, but that’s a constant
at some level, no matter what. Now you probably need
to have a little higher skill level at this stage of the game,
so you’re probably paying a little more, but certainly,
you’re not going out and approving a Teradata cluster.
You’re talking about Tier 3 storage. You’re talking about
a very low level of cost for the storage.”
--Bud Albers of Disney
PwC CTI
http://www.pwc.com/techforecast
Slide 17
Case study
Disney’s Hadoop cluster and central logging
service
Disney, 2010
PwC CTI
http://www.pwc.com/techforecast
Slide 18
Case Study
Yahoo: Expanding its use of Hadoop/NoSQL
“Soon after the Yahoo Search team started using Hadoop,
other parts of the company began to see the power and
flexibility that this system offers. Today, Yahoo uses Hadoop
for data warehousing, mail, spam detection, news feed
processing, and content/ad targeting.”
--Amr Awadallah of Cloudera
Some companies are now using NoSQL databases
in mission-critical applications
PwC CTI
http://www.pwc.com/techforecast
Slide 19
Devil’s advocate discussion #2
The stuff they process in a Hadoop cluster has so much noise in it that’s
it’s pretty useless for our purposes.
Open source data storage and databases seem relevant to web
companies, but not to most other industries. There’s no way we’d use
Hadoop in a data warehousing application.
Our company doesn’t have much of a development culture, and doing
lots of internal development has its own problems. I can’t see us
contributing code to a community.
PwC CTI
http://www.pwc.com/techforecast
The three waves of analytics
PwC CTI
PwC
http://www.pwc.com/techforecast
21
What’s really “new” about analytics?
More computing
speed, storage,
and ability to
scale
leads to
• More time and
better tools
• More data sources
• More focus on key
metrics
• Better access to results
leads to
A broader culture
of inquiry
Less guesswork u Less bias u More awareness u Better decisions
PwC CTI
http://www.pwc.com/techforecast
The three waves of analytics and a culture shift
PwC CTI
http://www.pwc.com/techforecast
Self-service visualization: What’s behind it
Source: Chris Stolte, Diane Tang, and Pat Hanrahan, “Computer systems and methods for the query and visualization of multidimensional databases” United States Patent
7089266, Stanford University, 2006, http://www.freepatentsonline.com/7089266.html, accessed February 12, 2012.
PwC CTI
http://www.pwc.com/techforecast
In-memory: Starts with BI, but doesn’t end there
Goal: straddling the boundary between
the core and the data warehouse
PwC CTI
http://www.pwc.com/techforecast
Social analytics: How NLP and graph analytics
helps
PwC CTI
http://www.pwc.com/techforecast
Big data and emerging cloud analytics (1 of 3)
* Churn: the proportion of contractual subscribers who leave during a given time period.
Source: Metamarkets and PwC, 2012
PwC CTI
http://www.pwc.com/techforecast
Big data and emerging cloud analytics (2 of 3)
PwC CTI
http://www.pwc.com/techforecast
Big data and emerging cloud analytics (3 of 3)
PwC CTI
http://www.pwc.com/techforecast
Devil’s advocate discussion #3
Social media intelligence is an oxymoron. Sentiment analysis, for
example, is misleading and often just plain useless.
You can give business units an analytics tool, but they’ll tend to misuse
it. The only worthwhile analysis comes from trained experts.
Sure, data services are emerging, but it’s very hard to trust what they
give you.
PwC CTI
http://www.pwc.com/techforecast
Other questions?
Alan Morrison
Center for Technology & Innovation
+1 (408) 817-5723
[email protected]
@AlanMorrison on Twitter
© 2012 PricewaterhouseCoopers LLP, a Delaware limited liability partnership. All rights reserved.
PwC refers to the US member firm, and may sometimes refer to the PwC network. Each member firm is a separate
legal entity. This document is for general information purposes only, and should not be used as a substitute for
consultation with professional advisors.
PwC CTI
http://www.pwc.com/techforecast
31