CIRUS : A Cloud Infrastructure for Real-time Ubilytics (aka - - PowerPoint PPT Presentation

cirus a cloud infrastructure for real time ubilytics aka
SMART_READER_LITE
LIVE PREVIEW

CIRUS : A Cloud Infrastructure for Real-time Ubilytics (aka - - PowerPoint PPT Presentation

CIRUS : A Cloud Infrastructure for Real-time Ubilytics (aka ubiquitous big data analytics) Didier Donsez Universit de Grenoble LIG / ERODS P.N@imag.fr 1 21/06/14 D. Donsez, CIRUS, EclipseCon 2014 From Processing.org Thanks to Manfred


slide-1
SLIDE 1

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

1

CIRUS : A Cloud Infrastructure for Real-time Ubilytics (aka ubiquitous big data analytics)

Didier Donsez

Université de Grenoble LIG / ERODS P.N@imag.fr

From Processing.org
slide-2
SLIDE 2

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

3

Thanks to Manfred for the introduction of my talk

slide-3
SLIDE 3

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

4

Web X.0 (X > 3) Emerging ICT domains

  • Cloud Computing
  • Big Data Analytics
  • Internet of (Every)Things
  • Social Networks
  • Mobile computing
  • Crowd sourcing
  • Open data
  • ...
slide-4
SLIDE 4

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

5

Internet of Things (IoT)

Instrumentation Communication Mediation Decision Action Mining

phones robot RFID / NFC SCADA sensor nodes Industrial IoT (IIoT)

slide-5
SLIDE 5

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

6

21/06/2014 6

Big Picture of Internet of (Every)Things, Data and Services

Home Automation SOHO Smart Public Space Smart Building Industry 4.0 Smart Cities Urban Spaces Geographic Scale Network cell size

WAN MAN LAN WLAN WSN PAN BAN

slide-6
SLIDE 6

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

7

Big Picture of Internet of (Every)Things, data and services

slide-7
SLIDE 7

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

8

Internet(s) of Data, Things and Services

  • Internet of (Chatty) Things
  • Internet of Everything
  • Internet of People
  • Internet of My Things
  • Industrial Internet of Things (IIoT) : Industry 4.0
  • Fog Computing
  • Cyber-Physical Systems
  • ...
slide-8
SLIDE 8

21/06/14

  • D. Donsez, Intergciels IoT

9

What is Cloud Computing ?

  • On-demand computing

– IaaS, PaaS, SaaS – Public, Private, Hybrid, Community, User-Centric,

Souverain

  • Advantages

– Virtualization, TCO, Resilience, Elasticity, Energy

efficency, Big Data Analytics …

  • Drawbacks

– Confidentiality (Privacy ,Industrial properties, …) – Souverainety

slide-9
SLIDE 9

10

Cloud Services Models

Virtual/Physical Infrastructure (FaaS) Infrastructure as a Service Amazon EC2, ... Platform as a Service Google App Engine, Amazon Hadoop ... Software as a Service Saleforce, Steam, ...

Smart Green Grid

H2

IT cooling

slide-10
SLIDE 10

21/06/14

  • D. Donsez, Intergciels IoT

11

Cloud Computing : UbiCloud, Cloud of Things, ...

  • UbiCloud

– Clouds with/for Ubi-terminals (smartphones, tablet

cars, IDS, ...)

  • Cloud of Things (CoT)

– Cloud for Things (data collection and long-term

storage ...)

– Things are facilities in the FaaS

slide-11
SLIDE 11

21/06/14

  • D. Donsez, Intergciels IoT

12

Cloud of Things

Virtual/Physical Infrastructure (FaaS) Inrastructure as a Service (IaaS)

(on-demand cpu/storage/nw infrastructures)

Platform as a Service (PaaS)

(on-demand platform of delivering your own application)

Software as a Service (SaaS)

(on-demand access to any applications)

Smart Green Grid

H2

IT cooling

Deltadrone, Cloud Robotics Xively, Axeda, Eurotech

slide-12
SLIDE 12

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

13

What is Big Data (Analytics) ?

slide-13
SLIDE 13

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

14

Big data is like teenage sex

  • “Big data is like teenage sex: everyone talks

about it, nobody really knows how to do it, everyone thinks everyone else is doing it, so everyone claims they are doing it…”

  • Dan Ariely, Professor at

Duke University, TED speaker

slide-14
SLIDE 14

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

15

The Data Deluge and the next IoT Data Deluge

http://www.snia.org/sites/default/files2/ABDS2012/Tutorials/RobPeglar_Introduction_Analytics%20_Big%20Data_Hadoop.pdf

1ZB=10^12 GB

slide-15
SLIDE 15

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

16

The 4+1V of Big Data

Volume Velocity Variety Veracity + Value

slide-16
SLIDE 16

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

17

IoT Big Data 5V

Volume Velocity Variety Veracity Value

slide-17
SLIDE 17

21/06/14

  • D. Donsez, Intergciels IoT

18

Big Data and IoT

  • Ubilytics : Ubiquitous big data analytics
  • Realtime prediction on the sensor data flows
  • For realtime decision (ie action)
  • Mixin with other data sources
  • Corporate data
  • Open Data (gov, ...)
  • Crowd-sourced data
  • Social networks posts/tweets
  • ...
Voir Talk ICAR 2013 : Big Data par Jean-Laurent Philippe, http://erods.liglab.fr/icar2013/programme.html#intel
slide-18
SLIDE 18

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

19

How analyze the Iot Data Deluge ?

  • Fastly (hours) and Very Fastly (milliseconds)
  • For speeding and improving decision supports

→ Business Intelligence tools (OLTP, OLAP, …) can't ! Now computing models are avalaible

  • Massively distributed, on-demand, fault tolerant
  • But
  • All old-fashioned statistical/prediction methods must be

rethink

slide-19
SLIDE 19

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

20

Computing Models for Big Data

  • Post-Processing Batch

– TBs / PBs of stored data

  • High-latency Decision Support

→ Map Reduce

Hadoop, SciDB, Spark, Giraph, ...

  • Continuous Event Streaming

– 100 MB/s of live data

  • Low-latency

Decision Support → Event Stream Processing

– Storm, S4, Samza, Millwheel, ..

  • Discretized Stream

Processing

  • Spark Streaming
  • Map-Update
  • MUD8P
slide-20
SLIDE 20

21

Event Stream Processing

  • Massively distributed processing of continuous

flows of events (sensors data, ...)

– Low-latency (few millisec after)

mutable state node 1 input records node 2 input records mutable state node 3 node 5 Node 4

  • utput

records

slide-21
SLIDE 21

22

Example : Event Stream Processing

Trending Topics

VoIPSTREAM (VS) Twitter Sentiment Analysis (SA)

From Maycon Bordin's ms thesis

slide-22
SLIDE 22 http://jameskinley.tumblr.com/post/37398560534/the-lambda-architecture-principles-for-architecting

Lambda Architecture

Nathan Marz (Twitter, Backtype)

  • Combine Low and High Latency BD stacks

Batch layer can compute the analytics model

  • f the Speed Layer

A n a l y t i c a l M

  • d

e l

slide-23
SLIDE 23

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

24

Ubilytics Ubiquitous Big Data Analytics

  • Motivation

– PaaS for « Ubilytics »

  • Autonomic : scalability, fault tolerance

– End-to-End

  • From sensors, gateways and lambda architecture (cloud)

– « Simple as Possible »

  • → for IoT SMEs & their IoT data scientists
  • Problem

– Huge variety of needs – Huge variety of technologies

slide-24
SLIDE 24

New trendy Job : IoT Data Scientist

http://nirvacana.com/thoughts/becoming-a-data-scientist/

Gartner says big data creates big jobs: 4.4 million IT jobs globally to support big d a t a b y 2 0 1 5 . http://www.gartner.com/newsroom/id/22 07915 The U.S. could face a shortage by 2018

  • f 140,000 to 190,000 people with "deep

analytical talent" and of 1.5 million people capable of analyzing data in ways that enable business decisions. (McKinsey & Co) Big Data industry is worth more than $100 billion growing at almost 10% a year (roughly twice as fast as the software business)

How can this guy deal with this deluge of technologies ? How to make this guy productive ?

slide-25
SLIDE 25

21/06/14

  • D. Donsez, CAL & CIEL 2014

26

Who is able to develop/ deploy Ublilytics infrastructures ? IoT Data Scientist

Cloud Computing Big Data Analytics Internet

  • f Things
slide-26
SLIDE 26

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

27

Towards "Ubilytics" PaaSs

Message Broker

  • r MaaS or PSaaS

Mosquitto, RabbitMQ, … Protocols : MQTT, AMQP, STOMP, XMPP, CoAP, WebRTC, Motwin ...

Message Broker

  • r MaaS or PSaaS

Mosquitto, RabbitMQ, … Protocols : MQTT, AMQP, STOMP, XMPP, CoAP, WebRTC, Motwin ...

SmartPhone

@ Car, City, …

M2M Gateway

@ Home, Office, City, Warehouse (OpenHAB, IoTSys)

Sensors data messages ie energy Consumption, temperature, images , ... Send selected sensors meausrements

Embedded boards smartphones by millions

In elastic Cloud

Realtime ESP Storm, Samza, S4, Spark Streaming, … topologies MQTT, AMQP, STOMP ... Monitoring Placement (static,dynamic)

Deployment & (Re)Configuration (Roboconf) MapReduce

Hadoop

NoSQL Store MongoDB, Cassandra, HDFS

In Elastic Hybrid Cloud

Computed prediction model

Predictions Trends, ...

Mashup, Reporting, Dashboard, . (history charts, ...)

Storing agregates

slide-27
SLIDE 27

28

Ubilytics Example Energy Consumption Forecast

Domain : Smart Grid

  • 2125 individual smart plugs in 40 houses
  • measuring and sending instant load (W) and cumulative load (kW)

Sensor Dataset*:

  • Events contain instant load (W) and cumulative load (kW)
  • 130 millions events/day (on one month)
  • 3 GB/day

Challenge* : Forecasts loads at 1min, 5min, 15min, 60min and 120min Goal : anticipate electricity demand for ajusting the production (ie. save energy and avoid blackout)

* DEBS GC 2014 http://www.cse.iitb.ac.in/debs2014/?page_id=42#

slide-28
SLIDE 28

29

Ubilytics Example Energy Consumption Forecast

instant load (W) and cumulative load (kW)

MQTT Broker + Storm Topology + Cassandra DB

  • n a Azure VM cluster
slide-29
SLIDE 29

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

30

Smart Campus (ie Small Smart City)

OpenHAB, Galileo, MQTT, Storm, Azure, ...

slide-30
SLIDE 30

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

31

Smart Campus

slide-31
SLIDE 31

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

32

Smart Campus

slide-32
SLIDE 32

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

33

SmartCampus

slide-33
SLIDE 33

21/06/14

  • D. Donsez, CIRUS, EclipseCon 2014

34

Smart Campus

slide-34
SLIDE 34

21/06/14

  • D. Donsez, CAL & CIEL 2014

35

Q Q &

& A

A