IBM Canada: Systems of Insight

© 2015 IBM Corporation

Cloud Data ServicesSystems of Engagement & Systems of Insight

Andrew BucklerWW Technical SalesMay 26th, 2015


� Mobile is the new battleground for business

� Web/Mobile architectures are different in kind than on-premises apps

� IBM provides best in class solutions for the three key components of a

modern, mobile, data-centric application:

Building a Modern, Data-Centric Application

The Platform: The Data: The Insight:


www.bluemix.net

www.cloudant.com

SDPSchema Discovery

Process

DataWorksData Refinery

Services

Cloud-Based Systems of Engagement(NoSQL, Mobile Apps, Internet of Things, Social Media)

IBM & Third Party Integrations(Cognos, SPSS, SAS, Tableau, ESRI ArcGIS)

Read/Write(HTTP)

Write

Read/Write

Read/Write

Read/Write(On/Off Prem)

Bluemix Platform as a Service

Systems of Record, Insight, & Engagement

www.dashDB.com© 2014 IBM Corporation

Systems of Record

Systems of Insight


Cloud Data Services Span Engagement, Insight, and Record

Enterprise Hadoop• Mission critical performance

• Automatic setup

• From development to business analytics

• Built for Systems of Insight

Cloudant DBaaS• Global data distribution

• Massively scalable

• Built for mobile, Systems of Engagement

dashDB• Columnar, in-memory performance

• Realtime query performance

• Data mining + geospatial

• Built for Systems of Insight & Record

SQLDB• Same skillset and data formats as

on-premises relational databases

• Built for Systems of Record

ANALYTICAL TRANSACTIONAL

UNSTRUCTURED

STRUCTURED

Mixed workloads and data types are knit together with DataWorks for true hybrid services


Services

Insight

Engagement


IBM Cloud

Beacons

MDM

Purchase Gateway

Mobile App Cloudant

dashDB

+ Watson

Business

Analysts

On-Prem

Systems of Record

Mobile Ecosystem


� Innovator of “Database as a

Service” (DBaaS)

� JSON document database

� For mobile, Web, & IoT apps

that need:– Elastic scalability

– High availability

– Data model flexibility

– Data mobility

� Available as:– Fully managed DBaaS

– On-premises private cloud

– Hybrid architecture

Build More. Grow More.

Sleep More.

Cloudant: Database as a Service for Systems of Engagement


Cloudant Combines the Best of Open Source

Engineered for DBaaS

Linear scalability, 24x7 uptime, Data durability, Enterprise security

Flexible development, Great ROI

Graphite &

RiemannMonitoring

ChefConfiguration

Management

rsyslogFederated Logging

CollectDContinuous

Integration

CouchDBJSON storage, API,

Replication

DynamoClustering, Scaling,

Fault Tolerance

LuceneText indexing &

Search

HaproxyGeo-Load Balancing

MongoDBDeclarative query

syntax

GeoJSONGeospatial indexing &

query


Digital safety deposit box web & mobile app for 20 million Fidelity

customers securely stores personal financial data

On-vehicle geospatial app for truck drivers optimizes routes,

guiding them to in-network refueling stations for lowest fuel prices

Cloud database for the US Intelligence community

On-line analysis of clinical trial data. Faster indexing and order of

magnitude less costly than RDBMS

Data layer for mobile gaming apps – 6 nodes to over 200 nodes in

1 year

Powers much of the Xbox One experience

“App store” catalog for Oculus VR Rift apps

Personal fitness mobile app with realtime mapping of running

routes; a top-5 fitness app in the App Store

Cloudant Use Cases Span Enterprise and Web/Mobile


dashDB: SQL Based System of Insight

� Deploy in minutes with rapid cloud provisioning

� No infrastructure investment for cloud agility

� Accelerate application development for analytics

� Deploy in minutes with rapid cloud provisioning

� No infrastructure investment for cloud agility

� Accelerate application development for analytics

� Built for Analytics to help you understand your data and business

� In-Database Analyticsfor greater efficiency and performance

� Compatible with Advanced Tooling like R and Watson Analytics

� Built for Analytics to help you understand your data and business

� In-Database Analyticsfor greater efficiency and performance

� Compatible with Advanced Tooling like R and Watson Analytics

� Grow more without growing the things that cost more

� Built-in Performance with in-memory technology

� Load and go withno tuning required

� Grow more without growing the things that cost more

� Built-in Performance with in-memory technology

� Load and go withno tuning required

Build More Know MoreGrow More

dashDB keeps data warehouse infrastructure out of your way

Data warehouse and analytics as a service on the cloud


The dashDB Cocktail: Three Parts IBM, ShakenD

BLU Acceleration

Netezza In-Database

Analytics

SoftLayer Infrastructure as a Service

Build More

Grow More

Know More

Focus on the business,

not the business of data warehousing!


dashDB Key Use Cases

• Extend on-premises data warehouse environments to the cloud

• Flexible, cost-effective growth

• Hybrid cloud models support ground to cloud

Extend / Modernize

• Easy synchronization of JSON to structured data

• Allows analytics via standard BI tools

• In-database predictive algorithms allow greater insight for Cloudant users than ever before

Cloudant Analytics

• Robust predictive analytic algorithms

• Integrated with R

• Watson Analytics ready

• Analytics ecosystem with partners

In-Database Analytics

• Data warehousing and analytics in the cloud

• Cloud agility and flexibility

• Analytics for cloud data, data marts, and development & test environments

Data Warehouse & Analytics Service

11



Services

Cloud Data Services Span Engagement, Insight, and Record

Enterprise Hadoop• Mission critical performance

• Automatic setup

• From development to business analytics

• Built for Systems of Insight

Cloudant DBaaS• Global data distribution

• Massively scalable

• Built for mobile, Systems of Engagement

dashDB• Columnar, in-memory performance

• Realtime query performance

• Data mining + geospatial

• Built for Systems of Insight & Record

SQLDB• Same skillset and data formats as

on-premises relational databases

• Built for Systems of Record

ANALYTICAL TRANSACTIONAL

UNSTRUCTURED

STRUCTURED

Mixed workloads and data types are knit together with DataWorks for true hybrid services


A new source of insight for retailers

DEMO

IBM Confidential © 2015 IBM Corporation

CDS Retail DemoMay 26, 2015Andrew Buckler – [email protected]


Overview

� The CDS Retail demo is meant to show Cloudant, dashDB, and Bluemix in the context of new mobile and analytics applications.

� The demo can be delivered at a high level to business people, or with more detail on the implementation for technical audiences.

� This PPT will shows how to build and execute the demo, and how to deliver the narrative/pitch behind the demo.

15


Setting up the Demo

� The demo is built from Cloudant, dashDB, Bluemix, and R. You will need accounts at cloudant.com and Bluemix to build the environment.

� Sign up for Cloudant: https://cloudant.com/sign-up/

� Sign up for Bluemix: https://apps.admin.ibmcloud.com/manage/trial/bluemix.html

� We start by replicating the demo’s data into your Cloudant system. The dashDB instance is then created from your Cloudant console. Finally, the R files are loaded into your dashDB instance’s R-Studio environment.

16


Replicating Cloudant Data

� Log in to your Cloudant console, then set up a new replication from remote database https://buckler.cloudant.com/retaildemo-store to a new local database called retaildemo-store.

Repeat for retaildemo-dept, retaildemo-desk, retaildemo-slot, and retaildemo-visitor

17


Replicating Cloudant Data

� Replication begins asynchronously and can take 5-10 minutes to complete. Validate your results with the images below.

18


Setting Up SDP from Cloudant to dashDB

� Setting up SDP into dashDB is very easy:

� Enter all 5 database names, your Bluemix credentials, then scroll down and click ‘Create Warehouse’.

19


Setting Up SDP from Cloudant to dashDB

� SDP has completed successfully when it looks like this:

20


Configuring the R Environment

� Click ‘Visit Warehouse’ to get to your dashDB console. Then click ‘Connect->Connection Settings’ to get the login/password for the RStudio console. These are the same login/password as the database connection, shown below:

21



� Now click ‘Analyze->R Scripts’ and then the ‘RStudio’ button. Use the login and password from your Connection Settings page to authenticate.

� The console is the pane on the left, where you will enter R commands. The file browser is in the lower right, where you will set up the prebuilt R/Shiny app.

22



� Enter these commands into the console:

install.packages( ‘ggplot2’ )

install.packages( ‘shiny’ )

� Then click ‘New Folder’ and title it ‘Retail’, open the retail folder and click ‘Upload’. Upload server.R and ui.R to this folder. The files are available in a zip archive from: https://w3-connections.ibm.com/files/app#/file/337f536a-23d7-4d62-9876-3cce53c324a2

23


Running the Demo

� You can start the web application by entering the following command into the RStudioconsole:

shiny::runApp( ‘Retail’ )

� A slider on the left controls time, and the checkboxes can be used to filter the visitors based on gender and loyalty program status. The three tabs on the right control which graphic is shown, either the map of the store, the spending share per department, or the time share per department.

� ALWAYS MOVE THE SLIDER TO 25+ BEFORE CHANGING TABS TO AVOID A BUG IN THE RENDERING CODE

24


Narrative

� The slide after this one is part of most dashDB decks. It describes the difference between systems of record, engagement, and insight.

• Systems of record are traditional IT systems like ERP, supply chain, payroll. These have primarily been built on-premises, and primarily are used to automate the business’s back office processes.

• Systems of engagement are new systems like mobile and sensor-enabled applications. These are primarily being built on the cloud, and are primarily used to automate the business’s front office processes.

• Systems of insight are the analytics that tie the data from these two types of systems together.

25

IBM Confidential © 2015 IBM Corporation26


Narrative

� The system of insight can mix engagement and record data to produce two beneficial outcomes for the business:

• Better operations: engagement data can show operational problems that the system of record data will miss.

• Better engagement: existing systems of record can be treasure troves of customer data, this data can analyzed and fed back to the system of engagement to produce better personalization and relevance to users.

• This demo is a simple example of how this can be accomplished.

27


Narrative

� Retailers are producing new mobile applications to act as their systems of engagement.

� These applications can detect beacons, small wireless devices that can be installed en masse within retail environments, and report the user’s location.

� Location data from the system of engagement is an entirely new source of insight into retail businesses, and this demo shows how simple IBM Cloud Data Services make it to take advantage.

28


Narrative

� The data set in the demo shows that shoppers are spending money primarily in department 5 (the Spend Share tab). This information is present in systems of record.

� However, shoppers are spending their time in department 2 (the Time Share tab). This information is only present in the system of engagement.

� Shoppers don’t spend time in a location without a reason, so they are either not finding what they want in Department 2 or the pricing is off.

� Only combining the two data sets in a system of insight lets the business discover the problem.

29


Narrative

� The type of mobile/location information used in this demo can be used in a variety of other ways:

Shoppers can be identified at the Point of Sale location even if they pay by cash and do not swipe a loyalty card. This deanonymizes the cash transaction and allows it to be associated with the shopper’s profile.

Shoppers can be detected upon entry and guided to new merchandise in the store that may appeal to them.

If an item is out of stock, a ‘clicks and mortar’ approach can use the mobile app to order and ship the item, preventing the shopper from buying from a competitor.

30


Technical Details

� It may be worth walking the client through how the demo is built, at a very high level. Start with Cloudant and show how easy it is to move data into dashDB.

� Show the dashDB console and how simple it is to browse data, then move to R to show how easy analytics can be.

� The ui.R and server.R files are small and simple, point out that a developer could produce the web demo in an afternoon.

� Reinforce that the web-based deployment, management, and development paradigm used by Bluemix and CDS allows the demo to be built without leaving a web browser.

31

32 #SmarterBiz

NEXT STEPS:

1. Visit me in the Social HubE Let’s Talk & Tweet #SmarterBiz

2. See our Analytics Representative in the Tech Showcase for detailed demonstrations

3. Please Sign Up for Watson Analytics: https://apps.admin.ibmcloud.com/manage/trial/watsonanalytics.html?watsonanalytics=true&cm_mmc=WAMicrositeOrganic--C24803SW&S_TACT=C24803SW

Cognos Express: http://www.ibm.com/developerworks/downloads/im/cognosexpress/index.html

4. For Additional Information

or type this URL into your browser: ibm.biz/sbs2015

IBM Canada: Systems of Insight

Technology

ibm corporation mobile

data durability

data formats

data types

ibm corporation cloudant

service systems of record

datacentric application

mobile apps