This is a continuation from Part 1. Why has the Internet of Things (IoT) become a big deal now?
When you follow the ICT field, you notice that when a new market starts to emerge, multiple companies jump in almost at the same time. This is easy to explain.
1. Progress makes existing technologies less expensive, more compact, more available, and therefore easier to apply.
2. A lot of smart people are constantly looking for a better new technology and can exploit such progress.
Why a small number of these companies win the market is a subject for another blog, and I do not touch it here. But Geoffrey Moore has a few books on this very subject.
Rezendes said that the following made IoT more realizable:
- Corporations and other organizations have come to rely more on data in making decisions.
- Connectivity is nearly ubiquitous.
- Technologies and components carry less technical risk and are more available.
- Physical deployment and logical integration are improved.
Areas of applicability
OK. What are the areas of IoT applicability?
Rezendes listed the following areas:
Incidentally, the Industrial Internet Consortium (IIC) deals with his points. IIC’s website states that it:
“was founded in March 2014 to bring together the organizations and technologies necessary to accelerate growth of the Industrial Internet by identifying, assembling and promoting best practices.”
The founding members are AT&T, Cisco, GE, Intel, and IBM. They focus on IoT in industrial segments, especially the following vertical markets.
Barriers and Hurdles
Rezendes discussed barriers and how they relate to the embedded community:
What the community can already handle and have an impact on:
What the community has an influence on:
What has impacts on the community’s business but the community has a limited impact on:
- Business models
Others also discuss some of the hurdles to IoT. Dr. Richard Soley, Executive Director, Industrial Internet Consortium, has mentioned a few roadblocks here. They are security, data privacy, technology, interoperability, and industry fragmentation.
Rich Quinnell of EE Times had a timely article that talked about the hurdles for IoT to overcome: 1) security, 2) standardization, 3) avoidance of siloed approaches, 4) adoption of data-centric design requirements, 5) development of hybrid business models, 6) device management, 7) power efficiency, 8 ) common development environment between cloud computing and embedded computing, 9) talent and expertise acquisition, and 10) data diversity.
Each point above will require a blog discussion of its own.
Rezendes also talked about fog computing, a term coined by Cisco and also known as edge computing. (If googled, the terms fog computing and edge computing yield 254,000 and 82,400 hits, respectively.) It is getting a lot of attention; a Fog Computing Conference was held in November 2014.
Because of the huge number of remotely distributed objects, it would make sense to adopt a new computing paradigm, in addition to cloud computing, for the reasons of latency, bandwidth, response time, and cost. Some cases, such as automobile collision avoidance systems, require real-time processing rather than sending data to the cloud, having them processed somewhere in the cloud, and waiting for the answer. The answer has to come within seconds; the need cannot wait for an action initiated by the cloud far away. Local IoT nodes should be able to make a decision based on the data sensed.
Note that cloud and fog computing complement each other, and fog computing does not replace cloud computing. Fog computing by itself deserves multiple blogs, but I won’t talk about it further here.
Rezendes’s one-hour talk stayed at the overview level, but it stimulated my interest in the industrial Internet of Things (IIT) and made me reconnect with Dr. Richard Soley, whom I knew when we were working on Object Request Broker (ORB) back in the early 1990s.
These were my takeaways from his talk:
- IoT is beyond SNS, and a “thing” can be any object in the physical world, including a person. He was talking about the Industrial Internet of Things (IIT).
- Unlike in SNS, a mechanism needs to be incorporated into such an object for monitoring, digitalizing, and transmitting information. Such a mechanism is an embedded system. The embedded community has a large impact on IoT and IIT.
I will cover this topic further in my upcoming blogs.