An Executive Summary of Internet of Things Technology Overview - Assembling the Technologies from previous blogs into Business Solutions
Vice President and Principal Analyst, Constellation Research
Constellation Research
Andy Mulholland is Vice President and Principal Analyst focusing on cloud business models. Formerly the Global Chief Technology Officer for the Capgemini Group from 2001 to 2011, Mulholland successfully led the organization through a period of mass disruption. Mulholland brings this experience to Constellation’s clients seeking to understand how Digital Business models will be built and deployed in conjunction with existing IT systems.
Coverage Areas
Consumerization of IT & The New C-Suite: BYOD,
Internet of Things, IoT, technology and business use
Previous experience:
Mulholland co authored four major books that chronicled the change and its impact on Enterprises starting in 2006 with the well-recognised book ‘Mashup Corporations’ with Chris Thomas of Intel. This was followed in…...
Read more
The Internet of Things, or IoT, is currently moving towards the top of the hype curve driven by a series of amazing predictions. However if asked to explain what IoT brings to business capabilities, or what technology components comprise an IoT solution, remarkably few people would be able to answer comprehensively. Those who did answer would tend to be able to identify just a single aspect of IoT; and what that was would be dependent on whether they worked in Operational Technology, Information Technology, or even Social Marketing. It’s fair to say that there is a lot of confusion about IoT, not helped by apparently conflicting views.
At the same point in the hype curve for Web technology equally few people would have been able describe the role that the ‘Internet of the Web’ would play in globalizing Content, creating mass participation of consumers, and laying the foundations for online ecommerce. Yet there are strong similarities in terms of the development curve and global impact with IoT.
The ‘Internet of Things’, IoT, makes use of Internet connectivity to add a new layer of capability around the ubiquitous Global scaling of ‘Interactions and Events’ between Intelligent Devices. This is a different layer with distinctly different functionality and technology to the ‘Internet of the Web’ with its focus what and how ‘Content’ can be used.
It should be noticeable that the above IoT definition that doesn’t refer to, or restrict itself, to the use of ‘sensors’ with simple event triggers, instead it embraces the reality of the numbers of devices and their characteristics that are, and will be, connected to the Internet. As an obvious example your Smart Phone working as your personal IoT device handing complex relationships between you, your location, your smart home, and IoT based services such as Uber Taxis.
It is best to define IoT by recognizing four important characteristics that between them are redefining capabilities, both for technology and the resulting business beneficial solutions.
- Intelligent; An addition to Moore’s law written today would probably dwell on how the cost of the same amount of processor power halves every eighteen months, with the result that many basic devices such as Coffee machines, or Hair curlers, now use processors to control their operation cycle. Size, Power, Cost and programming complexity are no longer barriers to adding intelligence to anything, the results are already clear to see.
- Connected; self-evident perhaps, but the physical formats by which connectivity is achieved have benefitted from the development of Mobility in all its forms. Today options range from low cost low data rate wireless connections that require no direct power through to high through put dedicated wired bus topologies. Taken with ubiquitous network types such as 4G wireless, or wired Ethernet, anything can now be connected at the right cost for the Service supported.
- Interactive; Intelligent connected devices no longer are limited to reporting simple sensed conditions, increasingly they participate with other devices, gateways, or cloud based Services, in a series of iterative interactions that will refine the details and context of an event into an insightful actionable outcome.
- Autonomous; As the sheer scale in the numbers of intelligent, connected, interactive devices grows so does the need for localized autonomous groups. As with the Web itself IoT is a decentralized environment, supporting scale through its decentralized loose-coupled architecture. For IoT centralization of all interactions and data is undesirable due several reasons including bandwidth availability, timeliness of response being impacted by latency. The required Autonomous groups may be pre-defined, or dynamically created; for IoT clusters where interactions and actionable outcomes have to be seen as ‘real time’. However it is likely that there will be an upstream outcome report made to other less time sensitive Services.
Time is a very key factor in creating architected IoT Smart Service business solutions based on these four functional, but the concept of Time is not directly comparable with the definition associated with data integration in Enterprise Architecture. Time to respond with an insight or an action is what defines the use of decentralized Fog Computing, or more centralized Cloud Based Services, and of course how any integration with legacy IT Enterprise Applications will be achieved.
As the numbers of connected devices continues to grow exponentially future competitive business success seems certain to depend of an Enterprise’s capability to outplay their competitors through using the four capabilities of IoT to optimize their dynamic tailored ‘Service’ led market responses. Smarter, faster, better connected and autonomously optimizing are all the wining traits in IoT based Smart Services.
Adding online Web based channels to increase sales within the current Business model is a survival necessity, but doesn’t provide a defense against first movers who use IoT based Smart Services to produce a disruptive competitive market impact.
For many Enterprises their first experiences with IoT will follow a similar path to their adoption of the Web starting with one, or probably more, initiatives on the internal Enterprise network. As with the Web adoption it is more likely that the earliest of these will be carried out by small internal groups of enthusiasts, and not necessarily known to senior management. The Web spread within the Enterprise as these initiatives became of increasing value for sharing previously isolated and inaccessible content. (It is as well to remember how quickly many Enterprises found themselves with an uncontrolled, accelerating number of individual Web sites, and decide to establish better managed adoption of IoT).
Connecting, and using data, from a variety of smart devices is already an established reality in factory and other process centric Industry sectors such as Oil & Gas under the heading of Process Automation and/or Operational Technology. Under the guise of improving interaction with actual, or potential, customers there has been a similar adoption in Social Marketing Interactions. Social Marketing use of IoT goes largely un-noticed until it is realized that much of the data being created and used is comes via the Smart Phone providing the four principles of an IoT Devices to ‘sense’ events and circumstances of the owner/user.
There is a substantive gap between those parts of an Enterprise that thrive on real time interactions and resulting data, versus IT operations driven by transactional applications and analysis of historic collected data. This is not surprising at this stage in the adoption curve as Business value is recognized from the edge of the Enterprise in its engagement with the market. Later with scale, and the recognition that it is now Business critical, the management of new technologies has in the past changed to the IT department.
The importance of time as a critical factor in the design of IoT Business solutions has already been raised. Time is also one of the major factors as to why different people in different technology and business areas don’t recognize a common IoT model. There are three separate definable and different time zone based requirements that can be present in IoT Enterprise level architecture. Although it is not necessary for all three to be present in every business valuable solution it is increasingly common with scale and in defining Business valuable Smart Services to find a chain reaction across all three Time zones.
When describing each zone it helps to illustrate their use with an appropriate example.
Quasi Real Time; IoT clusters, (fog computing environments) where alarm conditions require to as near as possible real time responses across a directly connected, often hard wired, group of Devices. This is common in factory machine management where an event starts with a sensor signaling a catastrophic failure needing immediate action. The event reaction may require adjustment to align with the exact state determined across the group of IoT devices (interconnected machines).
Quasi Real Time, and associated Fog Computing IoT clusters, are also prevalent in consumer IoT home based systems such as NEST, or HIVE, where the detection of an attempted home break-in immediately activates the home alarm siren, or a furnace leak turns off the water supply. Less noticeable, but even more highly developed are those used in new cars where the car itself is a Fog Computing IoT cluster able to sense and provide appropriate actions to a wide variety of Events. Due to connection constraints as well as the need for ‘immediate’ actions in some conditions, e.g., a tire blow out at high speed, a car is a perfect example of an Autonomous IoT environment.
An Enterprise should strategically seek to position as the ‘owner’ of Fog Computing environments that are align to its Business to ensure that the Processing of all Events initiates actionable Event responses related to their Business model. This is particularly relevant for any Manufacturing Business. It is not necessary to own all the IoT Devices indeed providing a Smart Service as connecting other IoT Devices increases the capability to provide unique value.
Notes; Quasi Real Time, (so called as exact Real Time cannot be achieved), is usually associated with Fog Computing; the creation of small edge based clouds, where the risk of network failure and the latency introduced by a remote cloud based Event Engine is unacceptable. A Fog Computing network is usually dedicated to the support of a single functional activity, using a specialized data protocol. Much of the so called ‘House Keeping’ data activity can be contained with in the Fog Computing IoT network to reduce the need to pass large amounts of data across low bandwidth edge of network connections. In scaled up Enterprise IoT Architecture Quasi Real Time consolidated key Event triggers would normally pass upstream to one, or more, Near Time IoT Service Clouds where it might be stored, or used to trigger a further level of responses.
Near Time; the upstream reporting point for Quasi Real Time IoT Events and the provider of Complex Event Processes for sophisticated multi Event Smart Services. A better name might be IoT Smart Services Clouds with the function to provide compute power and data storage at higher levels than in a Fog Computing cluster but with increased latency in response times. Smart Services Clouds would normally be distributed between multiple data centers to provide localized decentralized responses as timeliness and reliability remains key.
The most obvious, and widely appreciated, example of a Near Real time IoT Smart Services Cloud is Uber. Uber is not immediately recognizable as an IoT Smart Service, a point in common with many new Apps. In common with similar Apps it uses the four characteristics of IoT; intelligent end points; everything relevant is connected and autonomous; interactive optimization to achieve insightful actionable outcomes; and it happens in Near Time, (i.e. user perceived timeliness rather than time critical). It is an important principle of Near Time IoT Services, such as Uber, that failure is not critical, the event can be retried, together with relatively low amounts of data moving across the network as end points are on low bandwidth connections.
Notes; Near Time IoT usually requires a higher amount of Complex Event Processing from the Event Engine, associated with increased stored data from the Event Hub. In the case of Uber the action of asking for a Taxi is the creating event; the smart phone provides the intelligence as to location and personal details; the Near Time Service Cloud Event Engine compares this with the flow of data on Taxis types, locations, availability and makes an optimal match. The Uber Service Cloud contains other necessary connections such as data on traffic conditions and weather, some from Near Time Service Clouds, other data from Transaction Time, the third time zone. In the later category comes the integration for the Credit Card transaction as an example.
Near Time IoT Service Clouds will be the principle element in the development of high value Smart Services for the majority of Enterprises. The greater the number of Events and correlation with other inputs the greater the Business Value that can be created. Taken to the ultimate level Smart Services allow an Enterprise to disrupt an existing market by recombining the traditional elements into Smart Services that relate directly to the customers true requirement. E.g. Power by the Hour, Shared use of a Car schemes, etc.
Transaction Time; incorporates business outcomes that require some form of process action from Enterprise Applications such as Book to Bill financial recording in response to the use of a IoT Service. ERP, CRM and Data Analysis are all vital parts of Enterprise Back Office operations and their value is not diminished by the new business focuses brought by IoT. Similarly the output of existing analytical tools that can provide either a source of data to be included in an IoT Service Event Engine/Event Hub, or can be used to determine rules for Complex Event Processing.
As an example Uber uses weather data to dynamically adjust pricing in response to bad weather increasing demand for Taxis, an excellent example of dynamic pricing in response to real time conditions. Weather forecasting employs huge models with exceptional computational power, which run at set intervals. IoT Transaction Time relates to the updating of this type of data into the Near Time Service Cloud Event Hub data store. The triggering Event of calling for a cab results in the Event Engine adding this data to its Event Processing. Transaction Time IoT describes interactions with traditional IT systems for data outputs and/or inputs to existing systems and processes.
Selling the use of Smart Services requires the same Enterprises IT processes to be used as any other form of commercial transaction, and provides the same valuable input to Business Intelligence reporting. Conversely valuable contextual inputs can flow from Transactional IT to adjust the Event processing rules for the two other IoT time zones,
Representation of the Three Time Zones supporting IoT Smart Services
Resource
New C-Suite
Today the Institute of Industry Analyst Relations named Constellation Research 