The Future of Smart Systems
and IoT Analytics
The ability to detect patterns from device data
is the holy grail of smart systems
Scroll to access the full insight
Machine communications and the Internet of Things are combining to create new modes of asset awareness, intelligence and support services. In its simplest form, the Internet of Things is a concept in which inputs from machines, sensors, people, video streams, maps and more are digitized and placed onto networks. These inputs are integrated into Smart Systems that connect people, processes, and knowledge to enable collective awareness, efficiencies and better decision making. It is the combination of the data from machines with value added applications that delivers the value of the IoT.
We have now entered the age when everyday objects will communicate with, and control, other objects over networks—24/7/365. The objects are everything from consumer appliances and wearables, IT infrastructure and the elevator you’ve been waiting for; devices that enable industrial productivity, safe and comfortable indoor environments, consumer entertainment, health monitoring, safety, security and convenience to everyday life.
The convergence of large-scale data management and networked computing with real time machine intelligence is driving the integration of the physical and virtual worlds. The intersection of these trends – the Internet of Things and People – will create unimagined new values. Data management, analytic tools and new skills will be the core enablers of these new values.
Sensors, machines and a wide range of devices generate massive amounts of structured and unstructured data, requiring a whole new class of data modeling, management and analytics tools to uncover and capture value. In the hands of talented users, analysts, domain experts and data scientists, these data can generate productivity improvements, uncover operational risks, signal anomalies, eliminate inefficient service cycles, and even drive enhanced security protocols.
In a truly connected world of smart systems, not only people but all electronic and electro-mechanical products and machines will produce mountains of valuable information, all the time. Consider that today the number of connected devices on the planet has surpassed the number of people – 8+ billion – and depending on your definition of a sensor, there are already many more sensors on earth than people. This will generate phenomenal volumes of data ripe for value creation.
The ability to detect patterns from these devices is the holy grail of smart systems. Machine data analytics, often thought of as part of the evolving “big data” story, allows not only data patterns but a much higher order of intelligence to emerge from large collections of ordinary machine and device data. The implications of mining and analyzing machine data are immense; this is where the real core value creation opportunity lies within the Internet of Things.
DEVELOPING ANALYTICS APPLICATIONS THAT INTEGRATE DIVERSE DATA SOURCES AND FOCUS ON END USER VALUE WILL SEE BIG ADOPTION
In much the same way the HTML allows different browsers to accept and properly display a web document based on a known set of “markup rules,” Project Haystack is developing a standardized methodology and consensus-approved data models and tag libraries for diverse devices such as a temperature sensor or an electric meter. The essence of an “device” is completely abstracted from its real-world embodiment and is mutually interchangeable. The structure Project Haystack is utilizing is based on a simple model that includes entities such as:
» Site: single building with its own street address
» Equip: physical or logical piece of equipment within a site
» Point: sensor, actuator or setpoint value for an equip (the final element that data is associated with)
Each of these entities can then be tagged to capture essential information about its characteristics, purpose and relationships to other entities. It’s important to note that Haystack is cleverly combining a data modeling methodology with a community-developed tagging libraries (taxonomies) and the required communication protocols and APIs to exchange Haystack modeled data between applications and software tools, without them needing to be pre-aware of each other. This is a model we believe will be applicable in other industry sectors such as industrial systems. The open source nature of the project allows experts from any domain to build on the foundational work of the community to develop tagging models for systems in their own sphere.
SkyFoundry is leveraging this new semantic modeling approach to avoid the confinements and limitations of the today’s differing data types and relational database-oriented tools. It allows data to maintain their fundamental identity while bonding freely with other data. Facilitating discovery, based on data and information accessibility and cumulative systems intelligence, is one of the fundamental purposes of SkyFoundry’s platform. They are designing a system for a genuinely connected world in which there are no artificial barriers between pieces of information.
The SkySpark platform fundamentally changes the conventional paradigm, treating data from sensors, devices, systems and the physical world as “neutral” representations. In other words, treating diverse data types equally. This enables processes connecting diverse data in any combination to be rapidly built and deployed.