SCADA, factory automation, smart traffic control, connected telemetry devices, and many other forms of Machine to Machine communications have been in services for many years. So when we think about all of the hype around IoT it is a fair question to ask what has changed? Why do we believe that IoT will bring about the next “industrial revolution”, and what seismic shifts are in place to make this happen?
The real answer to this is that we have hit a “tipping point” of multiple technologies at once, which used together create industry changing opportunity. And this tipping point also comes at a time where there is a very real economic value in utilizing these technologies to create solutions to optimize industrial operations, enable new markets, and to create new value for consumers.
A brief look at the technologies that the technologies follows, first looking at the hardware side of the equation:
- The cost and power requirements for computation has dramatically declined while processing capabilities have dramatically increased. Low power, low cost devices are inexpensive and are enabling us to deploy smart solutions that would not have been practical before.
- Miniaturization – (which for silicon goes hand in hand with low power) has advanced to the point that processing solutions can be packaged in wearable devices, embedded unobtrusively into consumer products, and more or less invisibly into industrial applications.
- Sensor technology has continued to advance with MEMS technology and now gyroscopes, motion detectors, gas analysis, temperature, pressure, flow and a host of other detectors providing multiple insights to the physical world are also low cost, tiny, and low power.
- Communications silicon also continues to advance. While admittedly, more advances are needed to realize the full vision of IoT, many societies live in a world of WiFi and cellular ubiquity with easy connectivity to public and private transport cores and the internet. New wireless transport capabilities are emerging and competing to be the de facto standards for low power wide area network (LPWAN). Many of these technologies are available now and many are on the near-term horizon (WiFi Halo and LTE-M (LTE-MTC with DRX)).
- Also, while most prevalent on the software side, there is also an open source movement in hardware. With the maturity of Arduino, YUN, and Raspberry Pi platforms it is getting easier for developers – from hobbyists to industrial – to experiment with, develop and deploy IoT platforms. With this, silicon manufacturers find more competitive pressure to make it easier to deploy their products with working software and development tools to speed time to market.
- Out of the field, and in the enterprise or “cloud” side of the equation, the availability of high performance and low cost servers is also a major factor as we look at the overall architecture for IoT. Load balanced, Low cost, high performance server platforms provide the back-end processing of the high velocity, high volume and high variety (3 V’s) for IoT data, processing and analytics. When combined with cloud services and scalable, highly reliable “compute on demand” we have in hand the back end systems needed to support IoT.
Taken together, we have indeed reached a tipping point on the hardware side. The software side of the equation is equally dramatic, and huge advances on several fronts contribute to the the enablement of this new IoT solution opportunity:
- One of the most critical advances is the open source movement. While it may not be obvious how much influence this has had, the availability of open source operating systems has had a dramatic impact across the entire IT industry. The Linux movement has not only provided the industry with a rock-solid operating system, but the collaborative approach of the open source community has transformed the way the industry works and the speed that new ideas move forward. Open source has become a competitive necessity in many fields as customers now resist “vendor lock in”, and this further fuels the movement (witness recent open source cooperation with Virtualization, Cloud Foundry and OpenStack). The “behind the scenes” online tools and infrastructure that the open source community has created – which may not be apparent to many – is also accelerating the rate of success of open source development. GitHub is a great example of this, and for those who don’t know, aside from being a software version control tool, GitHub is a widely used repository for open source code, where developers share solutions across an amazing range of applications. StackOverflow is another example, where developers can post questions about problems they have encountered and an army of experts voluntarily reply with solutions. On-line how-to video solutions abound on YouTube, and on-line tutorial sites are becoming the norm with integrated “try it yourself” tools embedded to accelerate learning. There are many more examples of the revolution taking place in the open source community and and the impact cannot be overstated. This movement is gaining momentum at an exponential rate, and this is at the core of our new tipping point. Many of the solution components listed below are “children” of this movement.
- Cloud computing software is probably the next biggest game-change enabling the IoT tipping point. The scope of this is also hard to overstate. Virtualization has transformed the internet and allowed us to use back-end server hardware as a service where we can turn up or turn down platforms as required for the job at hand (Platform as a Service, or PaaS). Virtualization has also allowed us to realize Software as a Service (SaaS), where we load balance “cloned” systems as needed for scalable applications. This can all be delivered on top of platforms such as OpenStack, providing Infrastructure as a Service (IaaS). To quote from Wikipedia: “OpenStack is a free and open-sourcesoftware platformfor cloud computing, mostly deployed as an infrastructure-as-a-service (IaaS). The software platform consists of interrelated components that control hardware pools of processing, storage, and networking resources throughout a data center. Users either manage it through a web-based dashboard, through command-line tools, or through a RESTful API. OpenStack.org released it under the terms of the Apache License.” Now cloud compute providers such as Microsoft, Amazon, Google, Rackspace and others combine IaaS with operating systems, big data tool chains, development tools and Service Oriented Architecture (SOA) middleware to enable Platforms as a Service (PaaS). Together, this greatly simplifies rolling out resilient and scalable cloud compute resources suitable for large, medium and even small IoT solutions anywhere.
- Another critical enabling component for IoT is the “Big Data” movement, and many suggest that IoT is Big Data’s “killer app”. Hadoop and the Hadoop File System (HFS), and the multiple “Big Data” solutions that have emerged enable high available, high reliability, highly scalable “data lakes” for IoT. Multiple new high performance database technologies such as Cassandra, Mongo, Flume, Spark take us to the high performance realm beyond what SQL (Structured Query Language) systems could handle and high performance interworking solutions such as Kafka provide near-real-time data delivery across multiple software system modules. Big Data technologies, along with the farms of virtualized hardware mentioned above provide the means to handle the velocity, volume and variety (3V’s) of IoT data and processing.
- A more subtle, but no less important element is the availability and broad functionality of software protocols both open source and from value add solution providers. From communications protocols, to security, to message busses, full protocol stacks are available on many operating system alternatives ranging from back end server Linux platforms to low power solutions running on tiny ARM or even microcontroller cores. The more or less ubiquitous availability of software protocol stacks allow us a much simpler path to deliver secure and reliable end to end IoT solutions with shorter development and test cycles.
- Back end analytics are another key piece of the equation. Without the ability to perform smart analytics and to improve our operations of IoT devices, or to gain new insights about our businesses, or more generally, to operate our world better, there is really no point to IoT. But, the big data movement has brought significant investment to the analytics market, and multiple software solution providers are stepping up with IoT solutions.
- Finally, a rich ecosystem of web development tools, plugins, graphic utilities combined with a mature software development industry makes it easier than ever to present IoT data, analytics and human interfaces to enterprises, mobile or field personnel.
With the hardware and software advances highlighted above, our industry is well positioned for this seismic shift in capability called for with the IoT movement. With all of this momentum, we have critical mass and have reached this tipping point enabling this IoT era. Still, as an industry we have a lot of work ahead of us to make IoT easier and to simplify IoT deployments. We have great technology to put together great IoT solutions, but this industry is young and fragmented, and it remains way too hard for customers to build connect all of the pieces of end to end IoT solutions and deploy new applications. Gaps in standardization, integration of complementary layers of IoT solutions and the multiple competing alternatives possible can be daunting to anyone working to deploy solutions. Despite this fragmentation, we are well positioned to develop this industry and make custom end to end solutions much easier to roll out. This is a big focus in the IoT deployment sections of our work.
So with the above, and the fact that there is a lot of work to do notwithstanding, from a technology perspective we are at a real tipping point now and ready for IoT and this new world.
But, just because we make IoT happen, the most important thing to understand are the economic drivers of IoT. The economic value of IoT applications for consumers, enterprises, industry or “social good” projects have to be the foundation of IoT deployments are the real litmus test. This addressed in the link below.