Real-time Streaming Analytics: Business Value, Use Cases and Architectural Considerations: Impetus Webinar

7/22/2019 Real-time Streaming Analytics: Business Value, Use Cases and Architectural Considerations: Impetus Webinar

1/41

Real-time Streaming AnalyticsBusiness Value Use Cases andArchitectural Considerations

Big Data Solutions and Services Partner for Enterprises

Anand VenugopalSr. Director Business DevelopmentBig Data Solutions

1

Yue Cathy ChangSr. Director Business Development

Alliances and Partnerships


2/41

Picture your house

2


3/41

Whatif this was happening

nowto yourhome ?

3 Recorded version available at http://bit.ly/1i6OrwR
http://bit.ly/1i6OrwRhttp://bit.ly/1i6OrwR


4/41

Whendo you want to know ?

4

Later

or

Now ?

Recorded version available at http://bit.ly/1i6OrwR


5/41

Your best buddy from school

You havent met in years

5Recorded version available at http://bit.ly/1i6OrwR


6/41

Is in Vegas same time as you

Yourbuddy

You



7/41

Whendo you want to know ?

After you return

or

NOW ?



8/41

Whether individual or Business

Important things are always happening NOW

NOWis the ONLY time life REALLY happens

Maximize data valueprocess and actsooner!

Life is happening NOW

Real-time insight preserves or

creates value8


9/41

Topicswe will cover today

Why ?Business Value

What ?Is Real-time Streaming Analytics and what it isnot

How?Architectural considerations

Use casesWho and Where ?

What next ?Recommendations



10/41

What it is NOT

Quick-response interactive analytics on static data

Batch processing

Could be close but still NOTMicro Batch processing

What is Real-time Streaming

Analytics ?



11/41

Data analyzed in motionas it arrives

Routine: Monitoring, Counting, Alerting, Reporting

Complex Decision Making with Predictive analytics

Every incoming event is distinctly processible

Receive, Inspect, Analyse, Store, Distribute

Events may be stored later or in parallel

Immediate actions possible after processing

What is Real-time Streaming

Analytics ?



12/41

Realtime vs. Batchanalytics

SENSE Days ANALYSEWeek

sACT

SENSE ANALYSE ACT

Sec/

msSec/

ms

BATCH

Real time



13/41

13


14/41


Why ?Business Value







15/41

Business Value

Diminishes with the age of data

The drop is non-linear

$$$ ?Befor

e

Predictive analytics based on currentevents

Value depends on accuracy

$$NOW

Real-time

Certainty is high REAL

Value based on quickresponse

$$$Later

Descriptive

Diagnostic Least

value

15

ValueofData

Age ofData


16/41

Routine business operations (Real time systems)

Cutting preventable losses

Finding and monetizing missed opportunities

More revenue

Cost savings

Creating new opportunities

New Business models (Products, Services, Revenue)

Business Value from RTSA



17/41


Why ?Business Value







18/41

Routine Operations(RT systems)

ManufacturingControl Systems (Closed loop)

IT - Systems & Network Monitoring

Field Assets Monitoring and Alerting

Trucks, Oil rigs, Vending machines, Radio towers

Financial Transactions Processing

Authentications, Validations, Fraud



19/41

Cutting PreventableLosses

MH 370Loss of lives and assets

GMManufacturing defects

TargetMajor Security breach

Stock Exchange Meltdown

Many headline stories are failures inroutine operations and were preventable losses



20/41

Cutting PreventableLosses (2)

Medical / ClinicalComplex analytics in ICU

Disaster Warning Systems: Chile / Sandy

Brokerage - Fraudulent or Risky Trades

Preventive MaintenanceMachines, Plants

Customer Churn

Brand Reputation on Social Media



21/41

Missed Opportunities- Revenue

Customer Service always happens in real-time

Listening and Learning from customers (Social)

Context sensitive inventoryProducts, Ads

Recommend - UpsellCross-sell

SENSEArrival

ANALYSEContext

ACTServe



22/41

Missed Opportunities- Efficiency

Operational Efficiency of systems or processes

Network Optimization for cost, quality of service

Dynamic capacity management

Dynamic re-routing of traffic, cargo

Insurance AdjudicationDrone image analysis

SENSE

Operation

ANALYSE

Efficiency

ACT

Change



23/41

New Opportunities

Tractors are becoming soil sensors

Information service to farmers

Nikebecoming a healthcare company ??

Quantified self movement

Telecom giants selling data and insights

SENSE

What else ?

ANALYSE

What else ?

ACT

What else ?



24/41

Business Value of RTSASummary



25/41


Why ?Business Value


How ?Architectural considerations





26/41

ArchitecturalConsiderations(1/3)

SENSE

ANALYSEACT

APPROPRIATELY

ANALYSE

SENSE

ACT



27/41

Real time + Batch Analytics

27

Real-timeAnalytics

StreamProcessi

ng

BatchProcessin

g



28/41


28

StreamProcessing

Reliable ingest of multiple hi-volume hi-speedstreams

Route Data to Indexing and/or Storing DBs Distribution of data and analytics results Dynamic rule based routing



29/41


29

Real-timeAnalytics

Computations Rules-based Monitoring & Alerts Predictive Modelslive scoring of real

time data with run-time model refresh

StreamProcessin

g Hi-speed ingest of multiple hi-volume streams

Route to Indexing and Storing Distribution of data and

analytics results over ESB Dynamic rule based routing



30/41


30

Real-timeAnalytics



StreamProcessin

g Hi-speed ingest of multiplehi-volume streams

Route to Index and DB Distribution of data and

analytics results over ESB Dynamic rule based

routing

BatchProcessin

g

Receive, store and active archive ofdata-feeds

Offline analytics over ALL data Visual interactive BI/ Dashboards Offline building and updating of Predictive

Models for real-time scoring



31/41


31

Real-timeAnalytics



StreamProcessin

g Hi-speed ingest of multiple hi-

volume streams Route to Index and DB Distribution of data and analytics

results over ESB Dynamic rule based routing

BatchProcessing

Receive, store and active archive ofdata-feeds

Offline analytics over ALL data Visual interactive BI/ Dashboards Offline building and updating of

Predictive Models for real-time scoring



32/41

Real-time streaming

analytics pipeline and flow



33/41

Real-time streaming

analytics pipeline and flow

Scale and Robustness

Reliability -Guarantees

Publish-Subscribe

FlexibilityDynamicIntegration with Batch

Loose Coupling

Visualization

Ease of Administration



34/41

StreamAnalytix



35/41

Proprietary platforms

Vendor lock-in

No leverage of open source movement Do it yourself

Open source stitch up

Integration and maintenance nightmare Significant delays in time-to-market

Approaches to Stream Analytics



36/41

An App Server for real-time apps

Based on best-of-breed Open source

Focus on your Business logic leave infrastructure to theplatform

Handle all the 3Vs of Big Data in one platform

Seamless integration with Hadoop, NoSQL or any other DB

Rapidly operationalize pre-built analytical models or new ones

Significant time to market acceleration

Impetus provides full product support and professionalservices

Introducing StreamAnalytix



37/41


Why ?Business Value

What ?Is Real time streaming anaytics and what it is not






38/41

Important things are always happening NOW

Maximize data valueprocess and act sooner!

There is valuefind itImprove Ops, Cut losses, Findmissed & new opportunities

Architecture: SenseAnalyseActSense

RECAP

38

Real time insight preserves and

creates business value38


39/41

Get Real time streaming analytics in your roadmap

Talk to experienced peers and consultants

Start now with opportunities search, solution architecture andvendor conversations

Instrument (SENSE) everythingfind gaps and fill

Prove value with faster batch with current infra is possible

Establish mechanisms to ACT on the insights

Close the loopSense and Analyse effectiveness

DO IT

RECOMMENDATIONS



40/41


Why ?Business Value

What ?Is Real time streaming anaytics and what it is not




40


41/41

Big Data Solutions and Services partner for Enterprises

Request a demo of [email protected]

@impetustech
mailto:[email protected]:[email protected]

Real-time Streaming Analytics: Business Value, Use Cases and Architectural Considerations: Impetus Webinar

Documents