Last month in Monterrey I gave several talks about the how to make diverse apps in the Internet of Things (IoT) work together. It was an interesting crowd, with real problems in distributed telemetry (wastewater monitoring), big data (floor mat traffic analysis for retail) and missionary work. The problems ranged from few resource constraints to very tight constraints.

One of my favorite IoT Apps there was driven by missionary work, combining clean water Gospel messages in Africa. Both were powered by small solar PV installations, working within a tight budget of electrical power to provide fresh water and direct broadcast to cell phones. I am always amazed at how deep into the undeveloped world quite advanced phones have penetrated.

As the conference continued, I saw the case for my themes be made by the conference participants. By the end, I was seeing the resource framework developed for the US National Smart Grid effort everywhere. The challenge in that national effort was to improve the ability of the power grid and its successors to solve problems of supply and demand locally, and to do so in a manner that increases consumer choice, and accepts rapid technology innovation.

The solution then and now was micromarkets for power. Micromarkets have performed better than control systems in numerous published reports. Micromarkets support simple addition and removal of participation systems; they support very light integration. Micromarkets are insensitive to the processes embedded in technology diversity, and so can accept innovation.

In the national efforts, we went beyond power markets. Our communications specifications supported capacity markets, congestion markets, transmission markets, and even ancillary services markets. The complex Power Reserves market was modelled as a simpler options market. In each case, the market was a resource whose value was determined by time of delivery.

The components were time, product, and market services. For time, we developed communications for machine negotiation of human-centric schedules (WS-Calendar). For product, we developed abstract (an important point) models for describing products over time (EMIX), incorporating WS-Calendar. For market services, we developed the communication patterns necessary to negotiate for EMIX products.

What came to the fore was the importance of abstraction. EMIX was created without defining any particular product. We defined concrete types, meaning real products, for power and capacity and congestion, and so on. In Monterrey, we discussed the same communications for network bandwidth in data centers, for apps on limited cellular data plans, as well as for water-power combinations in sub-Saharan Africa. This month in Seattle, we will discuss using similar products to control storm water surges in wastewater systems.

[an error occurred while processing this directive]As the Internet of Things becomes the Internet of Doing Things, Apps will need to work within the boundaries of locally available resources. With more and more Apps, they will evolve to share these resources with each other. Integrating IoT applications through resource frameworks enables different Apps to share resources without knowing anything about each other. This “ignorance” is essential to rapid evolution of Apps in the IoT.

Resource markets are the simplest way to smooth interactions in a resource constrained world. In the IoT, there may be far more things than in traditional systems integration. Allocating budgets to participants in resource micromarkets may be the next thorny problem in the IoT. That leads to discussions of overlapping taxs and policy-based budgeting, subjects for another day.

I will be talking about resource frameworks this month at the AllSeen Alliance meeting in Seattle. I hope to see some of you there.

[an error occurred while processing this directive]
[Click Banner To Learn More]

[Home Page]  [The Automator]  [About]  [Subscribe ]  [Contact Us]