Operationalizing Data Analytics

1© Copyright 2014 Pivotal. All rights reserved. 1© Copyright 2014 Pivotal. All rights reserved.
Operationalizing
Data Analytics
How big data analytics frameworks
are evolving in the age of Hadoop
Jan 2015

2© Copyright 2014 Pivotal. All rights reserved.
Abstract
Hadoop is widely regarded as a key component of a Big Data infrastructure, but many companies have yet
to reap expected benefits from the platform.
In this Webinar, Brian Hopkins of Forrester, and Greg Chase of Pivotal examine the following:
Ÿ  Business-cases driving Hadoop deployments
Ÿ  Challenges translating deployment into business benefits
Ÿ  Other tools in a big data platform needed to realize insights
Ÿ  Future evolution of big data infrastructure with in-memory processing
Ÿ  How to achieve business value with data stored in Hadoop

Business cases
What is driving adoption of Hadoop?

How is Hadoop being used in the enterprise?
Hadoop has been considered key capability for implementing Big Data
Initiatives in the enterprise. What are the different ways Hadoop is being
used?
Ÿ  Hadoop adoption has been driven by large organizations with
lots of data and complex needs for insight
Ÿ  Early Hadoop adoption in large organizations was driven by
need to reduce cost of data persistence for analytics
Ÿ  Early use cases – Fraud, IT security, ad tech, pricing
optimization, offload transactional analytics
Ÿ  Emerging use cases – internet of things, customer journey path
analytics, dynamic digital experiences
Percentage of
firms that have
implemented
Hadoop
Percentage of firms
with > 20,000
employees that have
implemented Hadoop
Source: Forrester’s Business Technology Technographics Global Data And Analytics Survey, Q2 2014
Base: 1658 business and technology management professionals with knowledge of data and analytics

Hadoop Use Cases for Pivotal Customers
Retail
• CRM – Customer Scoring
• Store Siting and Layout
• Fraud Detection / Prevention
• Supply Chain Optimization
Advertising & Public Relations
• Demand Signaling
• Ad Targeting
• Sentiment Analysis
• Customer Acquisition
Financial Services
• Algorithmic Trading
• Risk Analysis
• Fraud Detection
• Portfolio Analysis
Media & Telecommunications
• Network Optimization
• Customer Scoring
• Churn Prevention
• Fraud Prevention
Manufacturing
• Product Research
• Engineering Analytics
• Process & Quality Analysis
• Distribution Optimization
Energy
• Smart Grid
• Exploration
Government
• Market Governance
• Counter-Terrorism
• Econometrics
• Health Informatics
Healthcare & Life Sciences
• Pharmaco-Genomics
• Bio-Informatics
• Pharmaceutical Research
• Clinical Outcomes Research

Challenges
What are typical barriers to achieving
business results with Hadoop?

How mature are data analytics on Hadoop, really?
Enterprises have invested in Hadoop but their analytics capabilities are still
at initial stages of maturity.
Ÿ  Enterprises are asking, “what do I need that I don’t have
already?”
–  Especially the larger ones that already have been investing for years in
analytics technology
Ÿ  Firms are struggling with business cases
Ÿ  Insight is elusive
–  Valuable insight is even more so
Ÿ  Experiment and fail fast
Large enterprises
say they have
enough big data
and are not
expanding
Source: Forrester’s Business Technology Technographics Global Data And Analytics Survey, Q2 2014
Base: 1658 business and technology management professionals with knowledge of data and analytics
Large enterprises
say their business
cases have a
proven ROI

Common Hadoop Analytics Challenges
1010101010101010101
1010101010101010101
1010101010101010101
Handling volatile
streaming data
1010101010101
1010101010101
1010101010101
Querying large
datasets
Applying advanced
analytics
10101010
10101
10101010
10101
10101010
In-Memory
Web
App
Web
App
Web
App
Data consistency at
scale

Realize Insights
How to apply advanced analytics to
data stored in Hadoop?

What they key data access methods?
What methods do you see enterprises taking to access and work
with big data?
Ÿ  The basic access paradigm is still KVS
Ÿ  MapReduce is too limiting for many cases – graph,
stream and SQL emerging
Ÿ  SQL/Hadoop is hot, and for good reason
–  Recognize the different flavors of this
–  It’s not an apples-to-apples comparison
Ÿ  Recognize the emerging need for speed – streaming and
in-memory
–  Accelerate mapReduce, graph, and search

A Business Data Lake Adds Analytic Insights
Centralized Management
System monitoring System management
Unified Data Management Tier
Data mgmt.
services
MDM
RDM
Audit and
policy mgmt.
Processing Tier
Workflow Management
Distillation Tier
HDFS storage
Unstructured and structured data
In-memory
MPP database
Unified Sources Flexible Actions
Real-time
ingestion
Micro batch
ingestion
Batch
ingestion
Real-time
insights
Interactive
insights
Batch
insights

In-Memory Computing
How will in-memory computing evolve
big data platforms?

What’s is happening with in-memory?
There is a lot of momentum around in-memory processing with
Spark and other technologies. What is really going on?
Ÿ  It’s important to understand the flavors of in-memory
–  Pure DBs, DB acceleration, caches/grids, Hadoop/Spark
Ÿ  Much of it is not all that new
–  Likely your DBs today are accelerated with better in-memory
caching
Ÿ  What is new/ interesting?
–  Spark (SparkSQL, SparkX, Spark Streaming, SparkML)
–  Spark will steal HDFS workloads
–  Streaming is a form of in-memory and comes in many flavors too
–  Caching

How In-Memory Evolves the Data Lake
Ÿ  Current: In-memory distributed databases: SQL & NoSQL
Ÿ  Future: Tachyon in-memory File System
–  Extending analytical data warehouses to in-memory OLAP
–  Convergence of in-memory OLAP and OLTP
–  Robust handling of Spark RDDs
Tez HAWQSpark GemFire
Tachyon (In-mem Polyglot File System)
HDFS NFS S3 Gluster FS
GemFireXD
Ceph

Business Value
How to achieve business value with
analytics based on Hadoop

How do enterprises cross the chasm?
How can enterprises get over the chasm in rapidly implementing
analytics (bottom up) and reaping business benefits (top down)?
Ÿ  Look for shared architecture requirements as you run
LOB specific pilots
–  Focus investment on metrics a LOB exec cares about
–  Demonstrate shared architecture benefits
Ÿ  Look for ways to automate and scale insights execution
–  Deliver insight at the point of decision
Ÿ  Stop using security as a blanket cloud objection
–  You likely already have sensitive customer data in the cloud
–  Leverage cloud when data is “soupy”

The Journey to Data Driven Innovation
STORE
Business Data
Lake
Store everything
ANALYZE
Big Data
Analytics
Generate Insights
BUILD
Data-Driven
Applications
Operationalize
INNOVATE
Agile
Enterprise
Iterate Rapidly
PDL Data
Science
Pivotal Labs
Agile
Pivotal CF
Services
PDL Data
Architecture
Agile Development
Big Data
SQL-Based Analytics
Enterprise PaaS

World’s Leading Experts
Pivotal Labs – Pivotal Data Labs
BATCH BATCH
INTERACTIVE INTERACTIVEHAWQGreenplum DB
Pivotal HD
REAL-TIME REAL-TIMEGemFire XDGemFire
The Foundation for Data-Driven Enterprise

Find out more…
Ÿ  Pivotal Big Data Suite
Ÿ  Download Pivotal HD
Ÿ  Enterprise SQL on Hadoop
Ÿ  Big Data @ Pivotal blog
–  The Future Architecture of the Data Lake

Thank You

Operationalizing Data Analytics

Recommended

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Operationalizing Data Analytics (20)

More from VMware Tanzu (20)

Operationalizing Data Analytics