Internet of Things
- 1. Internet of Things Aniekan Akpaffiong College Speaker Series January 2014
- 2. Introduction • Agenda – Internet of Things/Everything – Big Data – Cloud Computing – Virtualization
- 3. What Is Internet Of Things The general concept of the Internet of Things is that we can put a sensor on anything and have it send data back to a database through the Internet. In this way we can monitor everything, everywhere and build smarter systems that are more interactive than ever before. Dan Rowinski, article on ReadWrite.com
- 4. Google Trends Virtualization Cloud Computing Big Data Internet Of Things Graph generated 2014
- 5. Where Are The Things
- 6. How Many Things • Projections Vary – Uses cases still largely theoretical • Depends on what’s counted – E.g. include smart phones • Beware of motivations – Defining markets, setting agenda Example Estimates 2020 50 billion Cisco Systems, 2013 2020 26 billion Gartner
- 7. Internet of Things • “Internet of Things” – Popularized in 1999 by MIT research group – Sensor networks & content tagging to enable interaction with physical and logical objects – Core technologies • Nanotechnology, intelligent embedded systems, RFID, sensor technology,… – Platform technologies • IPv6, Big Data, Virtualization, Cloud Computing – Standards & APIs Entrance to SARAH, the artificially Intelligent "home of the future” in Syfy’s Eureka.
- 8. IoT Components • Enablers/Core – Wireless – ZigBee/NFC/RFID – MEMS/Nanotech – Content Analytics • Platform – IPv6 – Virtualization – Cloud Computing – Big Data • Objects Smart Lock, August.com
- 9. Internet of Things • Smart Planet (IBM), Planetary Skin (Cisco), CeNSE (HP),… • Connections, it’s not just of people anymore • IOT is about – Data – Sensors – Control – Analytics – Networks
- 10. Internet of Things • Applications of IoT – Manufacturing • Robotics, analytics, smart meters – Retail • Inventory tracking, – Healthcare • Remote monitoring, records – Transportation • Autonomous vehicles, GPS – Home • Monitoring, SMART devices, locks Thermostat, nest.com
- 11. Internet of Things • Concerns – Competence – Technocracy – Panopticon – Profiling – Hacking – Complexity – Inevitability – Data Ownership – Costs – Energy disney.wikia.com I’m not bad, I’m just drawn that way. – Jessica Rabbit
- 12. “If you have something that you don't want anyone to know, maybe you shouldn't be doing it in the first place.” – Google CEO Eric Schmidt, December 2009 CNBC interview Privacy & The Internet
- 13. Gartner Hype Cycle 2013 www.gartner.com/newsroom/id/2575515
- 15. Virtualization Hardware (CPU, RAM, HDD) Host Operating System App App Before Virtualization Hardware (CPU, RAM, HDD) Hypervisor App Virtual Machine #1 Virtual Machine #n Host Operating System 1 Host Operating System n App AppApp After Virtualization Virtualization is a methodology of dividing the resources of a computer hardware and software into multiple execution environments, known as virtual machines.
- 16. Virtualization Benefits Server Consolidation Power Savings Data Center Space Decrease Ease of use High Availability Simplified Management Improved Go-To-Market Time Enhanced Security Reduced Networking Costs Reduced Carbon Footprint Reduced TCO Desktop Consolidation Why People Love it Remember: Business cares about results, not virtualization
- 17. Virtualization Challenges Management Problem Isolation Charge-back and Ownership model Licensing Security and compliance Training 17 Why People Hate it
- 19. Cloud Computing Storage and Compute happens here Input and consumption of resulting information happens here Cloud
- 20. Cloud Computing • Computer scientist John McCarthy in the 1960s predicted that, "Computation may someday be organized as a public utility.“ • What is cloud computing? – In one of the more bare-bones definition, it is the ability to process information on someone else’s device. • Cloud computing essentially transfers computing tasks to the Internet.
- 21. “Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction” -- NIST The service is provided over a network as opposed to directly cabled. A broadband or high-speed network access is assumed. There are many use cases and many possible solution for cloud computing, any definition should be a model that can be modified to fit a particular use case. The resources have a management user interface allowing the user to configure and customize the resources as needed The asset or service can be requested (provisioned) more quickly than in the traditional IT model. Usually in seconds or minutes. The service provider should be able to “set it and forget it”. After the initial deployment, ongoing maintenance requirements should be minimal. This implies a self-service model. The user has direct access to the resources and can provision (request) and release (return) an asset with minimum overhead. The relevant compute resources are pooled and then logically divided and shared amongst the users. Related terms include multi-tenancy and elastic.
- 22. Cloud Computing – Deployment Models Cloud Types Properties Private Cloud For the exclusive use of a single organization. Can have multiple consumers, e.g. business units May exist on or off premise Community Cloud For exclusive use of a community with a shared concern or purpose. Owned and managed by the community and/or a third party. May exist on or off premise Public Cloud Available to the general public Owned and managed by a service provider organization. Exists on the premise of the service provider. Hybrid Cloud Composition of two or more distinct cloud infrastructures. Bound together by standard or proprietary technology yet otherwise distinct. Enables data and app portability
- 23. Big Data www-03.ibm.com/ibm/history/exhibits/storage/storage_PH0350A.html www.hoax-slayer.com/1956-hard-disk-drive.shtml Two things we do not see anymore today: 1. Pan-Am Airlines 2. The 1956 5 MB, 305 RAMAC HDD weighing approximately one ton
- 24. Big Data “There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy.” – Hamlet From the beginning of recorded time until 2003, we created 5 billion gigabytes of data. In 2011 the same amount was created every two days. By 2013 that time will shrink to 10 minutes. – The Human Face of Big Data – Rick Smolan & Jennifer Erwitt
- 25. SI Decimal Prefix Value Hard Drive Storage (decimal) Processor/Virtual Storage (binary) Binary Digit 100 1 Bit 1 Bit Byte 8 8 Bits 8 Bits Kilobyte 103 1000 Bytes 1024 Bytes Megabyte 106 1000 Kilobytes 1024 Kilobytes Gigabyte 109 1000 Megabytes 1024 Megabytes Terabyte 1012 1000 Gigabyte 1024 Gigabyte Petabyte 1015 1000 Terabyte 1024 Terabyte Exabyte 1018 1000 Petabyte 1024 Petabyte Zettabyte 1021 1000 Exabyte 1024 Exabyte Yottabyte 1024 1000 Zettabyte 1024 Zettabyte Brontobyte 1027 1000 Yottabyte 1024 Yottabyte Geopbyte 1030 1000 Brontobyte 1024 Brontobyte Binary Decision Text Character Half Page One Min. MP3 Audio 894,784 Plaintext Pages 4,581,298 books 268,435,456 MP3 Files 245 Million DVDs 375 Trillion Digital Pictures
- 26. What Is Big Data? • Big data – Too big in size, created too fast & with little or no standard structure – Structured + Semi-Structured + Unstructured Data – Social media, sensors, retail, weather, etc. – Difficult to process using traditional tools • Big data spans three dimensions: – Volume, Velocity and Variety.
- 27. The V’s of Big Data
- 28. Big Data Velocity Examples • In one second on the internet there are… – 197 Reddit votes cast – 463 Instagram Photos posted – 833 Tumblr posts posted – 1024 Skype calls made – 3935 Tweets tweeted – 11574 Dropbox files uploaded – 33,333 Google searches made – 46,333 YouTube videos viewed – 52,083 Facebook likes Captured in 2014
- 29. Big Data • Why Does Big data Matters? – New Data – Unlock Value – Shape The Future – Knowledge Is Power • Data Sources – Interconnectivity – Machines – Historical Image from commons.wikimedia.org
- 30. Analyzing Big Data • New Tools – Data Scientist – Scale-out Hardware – Parallel Programming Algorithms • MapReduce • Hadoop – Fast Storage: • In-memory • SSD • DAS – Languages for data mining/analytics: R, NoSQL, Pig/Hive/Hadoop, C/C++, Perl,…
- 31. Criticizing Big Data • Privacy • “Truthiness” • “Forest For The Tree” • Decisions based on the past • Correlation does not mean causation. • “Big Judgment” – Providing context for big data outcomes …not everything that can be counted counts, and not everything that counts can be counted. – William Bruce Cameron
- 32. Big Data Analytics “In a world where every click can be tracked and recorded, we shouldn’t be managing customers by putting them into groups of similar people, we really shouldn’t be guessing. We should be able to read the signals that customers are telling us to figure out what they want. I call that personalization.” – Nilan Peiris, CMTO HolidayExtras.com
- 33. Big Data Known Knowns • Things we know and we know that we know • working data Known Unknowns • Things we know that we do NOT know • data to be acquired Unknown Knowns • Things we do NOT know that we know • forgotten data Unknown Unknowns • Things we do NOT know that we do NOT know • data ignorance Popularized partly by Donald Rumsfeld
- 34. data engineer Data Information Knowledge Understanding Wisdom data engineer data analyst data miner data scientist raw data context experience models prediction The Data Scientist can obtain actionable insight from an analysis of the data. Big Data
- 35. Smart Dust The Sensors That Track Every Thing, Everywhere Sensors are ubiquitous, not just in devices
- 36. Thank You