ACC-2012, Bangalore, India, 28 July, 2012
1 like•2,719 views
The document discusses advancements in IoT, sensors, and big data, particularly focusing on the CSIRO's initiatives in sensing-as-a-service and cloud computing. It highlights the exponential growth of data from IoT devices and the challenges in managing this big data. The OpenIoT project aims to provide an open-source framework for IoT applications, enabling self-managing cloud environments and improving various sectors such as agriculture and disaster management.
![Mobile Crowd-sensing Applications
Environmental – monitor the
phenomena related to the natural
environment
pollution levels (Common Sense[1]),
water levels in creeks
(CreekWatch[2]), wildlife habitats
Infrastructure - measurement of
large-scale phenomena related to
public infrastructure
traffic congestion and road
conditions (CarTel [3], Nericell [4]),
parking availability (ParkNet [5]),
outages of public works (e.g.,
malfunctioning fire hydrants, broken
traffic lights),
real-time transit tracking](https://image.slidesharecdn.com/presentation7-130629140430-phpapp01/85/ACC-2012-Bangalore-India-28-July-2012-39-320.jpg)
![Mobile Crowd-sensing Applications
Health and well-being – monitor health and
well-being of groups of people
individuals measure location and bike route
quality (e.g., CO2 content on route, bumpiness
of ride), and aggregate the data to obtain the
“most” bikeable routes (BikeNet [6])
individuals take pictures of what they eat and
share it within a community to compare their
eating habits (DietSense[7]) – community of
diabetics
Social networking
user activity recognition and sharing with their
social networks (Cenceme[8])
social introduction /matchmaking based on user
profile information (Serendipity [9], WhozThat
[10])
group-based, context-aware recommendation of
music/videos (SocialFusion [11])
visualization of mobile user concentration in an
area using users’ GPS locations (Citysense [12])](https://image.slidesharecdn.com/presentation7-130629140430-phpapp01/85/ACC-2012-Bangalore-India-28-July-2012-40-320.jpg)
ACC-2012, Bangalore, India, 28 July, 2012
- 1. Arkady Zaslavsky, Charith Perera, Dimitrios Georgakopoulos CSIRO, Australia Sensing-as-a-service and Big Data Advances in Cloud Computing, Bangalore, 28 July, 2012
- 2. Outline 1. IoT 2. Sensors, sensor networks, sensing-as-a-service 3. Big Data 4. EU FP7 OpenIoT 5. CSIRO projects 6. Conclusion
- 3. 3 | CSIRO. Australian Science, Australia's Future 6500+ staff over 55 locations CSIRO today: a snapshot 170+ active licences of CSIRO innovation 150+ spin-offs based on our IP & expertise Ranked in top 1% in 14 research fields One of the largest & most diverse in the world Australia’s national science agency Building national prosperity and wellbeing
- 4. 4 | CSIRO. Australian Science, Australia's Future Future Manufacturing Light Metals Minerals Down Under Sustainable Agriculture Water for a Healthy Country Preventative Health Wealth from Oceans Climate Adaptation Food Futures Energy Transformed National Research Flagships
- 5. Future Internet Smart Grids Smart Meters Environmental Sensors Water Management Water Management Smart Agriculture IoT Media delivery anywhere New media types Many media managers Many media managers New ways of media consumption Media creation by everybody IoM Software as a Service Ambient services Cloud computing Green computing Hybrid cloud IoS Social media Social media mining Crowd sourcing Disaster management Telepresence & augment. reality IoP Cybersecurit y Compliance Trust IoE Future Internet Society
- 6. based on standard & interoperable communication protocols A dynamic global network infrastructure with self configuring capabilities are seamlessly integrated into the information network. virtual personalities, use intelligent interfaces, and where physical & virtual “things” have identities, physical attributes, Internet of Things IoT
- 7. Internet of Things Imagine a world where: • your car knows where the traffic jams / road anomalies are • your fridge knows how long before the milk expires • you know the areas with the less pollution where you can jog freely • energy resources can be managed efficiently And the list goes on forever!
- 8. Existing WSN Devices Wikipedia lists 31 Sensor Network devices Sensor Network Museum lists 30 devices http://www.snm.ethz.ch/Main/HomePage IMote 2.0: MicaZ: TelosB:
- 9. IoT is Already Here
- 10. The Internet Will Extend to Billions of Devices
- 11. Wireless and Sensors Will Be Everywhere 90% of world’s population has wireless connectivity 100,000 wireless network masts are erected annually 1 billion electronic devices equipped with WiFi shipped in 2012 Billions of smart sensors • Cisco2009: PlanetarySkin—integrate sensors on land, in sea, in air, and in space to help make it possible to see the whole picture when it comes to the effects to and changes in the environment • HP2010: Central Nervous System for the Earth‖(CeNSE) – 10-year mission to embed up to a trillion push-pin-sized sensors and actuators around the globe Source: Cisco IBSG, UN: International Telecommunications Union, Real-Aliens. Com 2006-2011
- 12. The “Zettaflood” is just the Beginning of the IoT Traffic Total IP Traffic on the global Internet: • 2003-1.8 Petabytes • 2007- 161 Exabytes • 2009- 487 Exabytes • 2010- ½Zettabyte • 2011- 1 ZettaByte (540,000 X increase from 2003) Expected to double over the next 18 months 2012- 91% expected to be video Source: VentureBeat, IDC, C|Net, TheGuardian, UK
- 14. Sensors: Technologies and Global Markets
- 15. Big Data CSIRO. Sensor Cloud and the Internet of Things The total amount of data generated on earth exceeded one Zettabyte (ZB) in 2010. It is predicted that data volume will grow exponentially as depicted (www.teradata.com)
- 16. CSIRO. Sensor Cloud and the Internet of Things McKinsey report, 2011
- 17. IoT & Big Data CSIRO. Sensor Cloud and the Internet of Things Data generated from the Internet of Things will grow exponentially as the number of connected nodes increases. Estimated numbers of connected nodes based on different sectors are presented in Millions
- 18. CHALLENGES IN BIG DATA MANAGEMENT High volume of processing using low power consumed digital processing architecture. Discovery of data-adaptive Machine learning techniques that can analyse data in real- time. Design scalable data storages that provide efficient data mining. CSIRO. Sensor Cloud and the Internet of Things
- 19. EU FP7 OpenIoT Objectives The Motivation: • Despite the proliferation of pervasive grids, participatory sensing and on-demand sensing services (e.g., «Location-as-a-Service») there is no easy (and generic) way to offer utility based IoT services • IoT end-users and providers still need to deploy their own sensors and devices The Open IoT Goal: • Research and provide an open source middleware framework enabling the dynamic, self-organizing formulation of self-managing cloud environments for IoT applications • Sensing-as-a-Service • Converge IoT and IoS - Cloud Computing www.openiot.eu
- 21. OpenIoT experimental test-bed For the High Resolution Plant Phenomics Centre’s Phenonet project • Measure environmental and plant physiology parameters in the field • Improve the quality and scale of data available to plant breeders from grain trial plantings IE Lab contribution • Design and programming of sensor network • Testbed for declarative programming of sensor networks • Fast browser-based data display & analysis using reusable components • See http://phenonet.com
- 22. Sensor information management solution Runs on base computers (Java) Integrates data streams with R/Matlab,… Publish/Subscribe with continuous data processing Many other advanced features… OpenIoT is extending and transforming GSN to a Cloud Service What is GSN ?
- 23. Acquire data from sensors, RFIDs, motes, cameras, and other information sources (anything that can produce data streams) Stream Processing Engine • Window-based processing • Example: Sliding Value 1, Window Size 3 Publish/Subscribe services Archiving GSN Focus
- 25. Some of the Existing Deployments
- 26. • GSN Applications are collections of interconnected Virtual Sensors • Example: • 9 Virtual Sensors • Each virtual sensor has to configure its data source • Publish/Subscribe Model • Continuous Queries Building GSN Applications
- 27. GSN Virtual Sensors Virtual sensor are specified by XML scripts Include stream data processing steps specified via SQL-like queries Include References to Java classes of additional operators, including – Configuration of data steam sources via provided wrappers – Non-relational stream data processing operators, e.g., for spatiotemporal processing, alerting, for data cleaning, visualization Source 1 Source 2 … Source n 1. Data source configuration • Reference to Java Classes 2. SQL-based processing steps 3. Non-SQL processing steps • Reference to Java Classes Output Stream <XML>
- 28. GSN Sensor integration support wrappers • GSN provides wrappers and APIs for adding more • Wrappers produce streams • Interact with sensors –Wrappers consume control cmds 20+ Wrappers already available • TinyOS, CSV, HTTP-Get, Serial Port, UDP, RSS Feeds, … An introduction on how to write a new wrapper • sourceforge.net/apps/trac/gsn/wiki/writing-wrapper • sourceforge.net/apps/trac/gsn/wiki/template-wrapper Sensor Wrapper Stream Control/ Feedback Flexible Data Acquisition for Virtual Sensors
- 29. CSIRO Applications CSIRO. Sensor Cloud and the Internet of Things
- 32. Priority areas SenseT Middleware Workshop | Tasmanian Applications | 26-27 April 2012 1. Circular Head – NW Tasmania Agriculture (dairy), aquaculture, carbon markets 2. Scottsdale – NE Tasmania Agriculture (dairy, fruit and viticulture), food logistics, carbon markets 3. South Esk – Northern Midlands, NE Tasmania Catchment management, flood prediction, carbon markets 4. Triabunna – East Coast Agriculture (pasture, fruit and viticulture), aquaculture, food logistics, carbon markets 5. Huon Valley – SE Tasmania Aquaculture, food logistics, carbon markets, catchment management
- 33. • Strategic relevance - Improve productivity of the salmon and oyster aquaculture industry through provision of real-time awareness of critical environmental variables. - Build supply chain resilience to natural events (e.g. harmful algal blooms) that may force temporary closure of some production facilities. • Impact - Tasmanian aquaculture industry groups (Oysters Tasmania and Tasmanian Salmon Growers Association) - Tasmania Department of Human and Health Services Aquaculture monitoring SenseT Middleware Workshop | Tasmanian Applications | 26-27 April 2012
- 34. • Science challenges - Federation of existing aquaculture data including sensor observations, historical data, and model outputs. - Develop incremental learning model to analyse cause-effect relationships. Decision support tools capable of predicting events that may potentially impact productivity. Aquaculture monitoring SenseT Middleware Workshop | Tasmanian Applications | 26-27 April 2012
- 35. • Strategic relevance - Real-time sensing and modelling of environmental and irrigation crop parameters to: - Increase water availability - Improve water use efficiency - Boost agricultural productivity without compromising critical ecosystem services. • Impact - Tasmanian Irrigation (Lower South Esk Irrigation Scheme). Smart rural infrastructure SenseT Middleware Workshop | Tasmanian Applications | 26-27 April 2012
- 36. • Science challenges - Federation/querying of of diverse sensor data streams. - Model-sensor integration, framework for managing computer simulation models used to predict weather, stream flow, crop production, and irrigation demand. - Provenance-based model-sensitivity analysis. - Decision support tools that will maximise irrigation water use efficiency and automate irrigation. Smart rural infrastructure SenseT Middleware Workshop | Tasmanian Applications | 26-27 April 2012
- 37. • Strategic relevance - Delivers on key objectives in the COAG National Disaster Resilience Statement. - Improves government and societal capacity to deal with disasters. • Impact - Tasmania State Emergency Services (SES). - Local communities. - Utilities. Context-based disaster alerts SenseT Middleware Workshop | Tasmanian Applications | 26-27 April 2012
- 38. • Science challenges - Alerting system is dependent on information from many different sources including live sensor feeds, spatial feature services, and output from predictive models. - Representing context information in a unified manner - Developing a methodology for analysing different contexts - Predicting causes and effects - Providing context-aware alerting services. Context-based disaster alerts SenseT Middleware Workshop | Tasmanian Applications | 26-27 April 2012
- 39. Mobile Crowd-sensing Applications Environmental – monitor the phenomena related to the natural environment pollution levels (Common Sense[1]), water levels in creeks (CreekWatch[2]), wildlife habitats Infrastructure - measurement of large-scale phenomena related to public infrastructure traffic congestion and road conditions (CarTel [3], Nericell [4]), parking availability (ParkNet [5]), outages of public works (e.g., malfunctioning fire hydrants, broken traffic lights), real-time transit tracking
- 40. Mobile Crowd-sensing Applications Health and well-being – monitor health and well-being of groups of people individuals measure location and bike route quality (e.g., CO2 content on route, bumpiness of ride), and aggregate the data to obtain the “most” bikeable routes (BikeNet [6]) individuals take pictures of what they eat and share it within a community to compare their eating habits (DietSense[7]) – community of diabetics Social networking user activity recognition and sharing with their social networks (Cenceme[8]) social introduction /matchmaking based on user profile information (Serendipity [9], WhozThat [10]) group-based, context-aware recommendation of music/videos (SocialFusion [11]) visualization of mobile user concentration in an area using users’ GPS locations (Citysense [12])
- 41. MCS Applications- Issues Require mobile devices to continuously sense, process and upload sensed data to the cloud/remote servers Data collection should be cost-efficient for both the devices and the networks Local analytics Require infrastructure to receive, manage and analyse large volumes of real-time data streams using cloud computing platforms Aggregate analytics Require participation from the user and willingness to allow collection of sensor data user preferences privacy incentive
- 42. Here-n-Now: A CAROMM application
- 43. How it works
- 44. Thank you ! Dr Arkady Zaslavsky, Professor Science Leader in Semantic Data Management, CSIRO, Australia Phone: 02 6216 7132 Email: [email protected] AND Research Professor Luleå University of Technology, Sweden CSIRO. Sensor Cloud and the Internet of Things