The Failure of IoT Platforms
CREATIVE EVOLUTION and
the POST PLATFORM ERA
When telephones first came into existence, all calls were routed through switchboards and had to be connected by a live operator. It was long ago forecast that if telephone traffic continued to grow in this way, soon everybody in the world would have to be a switchboard operator. Of course that did not happen because automation was built into the systems to handle common tasks like connecting calls.
We are quickly approaching analogous circumstances in the IoT arena with the proliferation of connected devices. The tools we are working with today to make products “smart” were not designed to handle the diversity of devices, the scope of interactions and the massive volume of data-points generated from devices. Each new device requires too much customization and maintenance just to perform the same basic tasks. These challenges are diluting the ability of organizations to efficiently and effectively manage development.
Today, platforms for the Internet of Things are still a kludgy collection of yesterday’s technology and architectures that do not address the most basic development challenges. Even though many companies are telling fantastic IoT marketing stories about what their solutions can do, you wouldn’t know it from today’s fragmented collection of incomplete platforms, narrow point-solutions, and software incompatibility.
We need better software to empower users and developers to exploit the vast potential of the Internet of Things.
PLATFORM OPPORTUNITY SCALE and DIMENSIONING
Through our firm’s consulting work with OEMs and services providers adopting new platforms and related technology, we have found many of our clients and others we speak with in the marketplace very confused about what actually is the definition of a so-called IoT platform and what capabilities and needs must be addressed by developers of platform technology.
The number of developers that have been funded to bring new IoT platform technology to market over the last five years is staggering. IT systems suppliers and wireless carriers have compounded the confusion by also introducing a variety of “platform” …“gateway”… “software” … etc… IoT offerings.
We decided we needed to analyze user needs and requirements more closely and then analyze the broad, diverse and confusing array of offerings in this context. As a result, Harbor Research screened over 200 software infrastructure and platform players to better understand each supplier’s core functions, enabling technologies as well the applications, use cases and markets served.
Demand for these types of capabilities is hardly new, but as they converge, these technologies need to be interwoven and mutually supportive. We believe success will only go to players who effectively leverage their combined potential.
Based on direct interviews with market participants and thought leaders, next generation IoT and Smart Systems platform architecture will need to address the following:
» Advanced, high value applications and use cases enabling systemic awareness, visibility and collaboration can be achieved using today’s technology, however, significant time and money is required to integrate and manage the variety of tools from numerous vendors.
» A new generation of systems architecture is required – one that can align with and “form fit” to the physical world; embedded computing with software tools that are easy to use and address a broad range of real-time, and historical data analytics requirements – ideally a single unified framework to design and build solutions that can interoperate across diverse data environments and under widely differing usage scenarios.
» Systems will require a true distributed software architecture that can process and create value from device data locally at the node – in remote sites, buildings, equipment rooms and telecom closets, while enabling higher level applications that generate value from portfolio-wide data. We have been lead to believe that the only way to get value from IoT data is to bring all of it to the cloud. It is not possible, cost effective or desirable to transmit every piece of data from IoT devices to the cloud in order to gain value from them. Numerous critical applications will require processing and analysis of data locally to support user needs and to automatically filter and semantically enrich that data before delivering it to higher levels of the architecture for other application needs.
» Users needs and requirements emphasize the need for software that can enable easy integration of any combination of inputs and data types – message – feed – stream – data – in real-time with stream processing that provides complete independence from traditional rigid database technologies; software designed without the bias and dependencies of a single product or service application that will cause integration pain when attempted to be used beyond its original scope.
» Ultimately, software and tools that allow users to easily and quickly build their own applications integrated into a truly scalable software architecture and data models/architecture with no constraints on where it gets deployed – whether on a chip, device, server, cloud, or hybrid system; yet able to provide extensibility with common features and functions at each level of the architecture.
» The rigid and fragmented nature of software offerings available today make it extremely difficult and expensive to develop effective Smart Systems and IoT applications. Feedback from developers, adopters and users suggests we are reaching a critical juncture in market development where organizations will expect their invested resources to develop new data-centric IoT applications can be easily re-used again and again. We believe this situation underscores the need for fresh new thinking about computing, information management and networking that demonstrates it is possible to migrate gracefully to scalable architectures truly designed for the era of pervasive computing.
EVOLVING PLATFORM PLAYER ECOSYSTEMS
After displaying signs of relative indifference for so many years, traditional players within established segments such as IT systems, enterprise software, telecom and network equipment have finally awakened over the last three or so years. Meanwhile, the world of Smart Systems and Services has continued to evolve, particularly in the emergent player communities serving this opportunity. In this new cycle, physical assets and devices are being connected and enabled with sensors on a daily basis. Recognizing this scenario, technology suppliers around the world are now devoting major resources to addressing the opportunities associated with Smart Systems, but are, for the most part, failing to bring any new fundamental innovations.
For most IT and telco equipment vendors, selling existing servers, PCs, storage devices, and enterprise applications will be a recipe for shrinking revenues, given the impacts of new architectures on these markets and applications. But demand for these purchased technologies will not disappear. A few vendors that focus on being the consolidators of these shrinking but still large product markets will be able to survive.
The functional requirements of new digital and IoT platforms increases with the complexity of the use cases as well as the scope of assets to be targeted; the IoT is not a simple environment to navigate. High value use cases enabling widespread visibility and collaboration can be achieved using today’s technology, however, significant time and money is required to integrate and manage the variety of tools from numerous vendors.
Who will be the winners and who will lose as the Smart Systems Platform opportunity develops? In the much larger and more complex world of business-to-business technology, there is no vendor that has the clear leadership position. While the “Internet of Things” platform opportunity represents a market of vast potential, technology suppliers must be aware of the current competitive dynamics if they are to successfully navigate the market. How well will various supplier groups align with the Smart Systems opportunity as it develops?
Players like HPE Dell/EMC, Lenova and other peer legacy IT systems players are all still in the horizontal IT tech space and have been slow to embrace Smart Systems in a meaningful manner beyond “point” maneuvers such as introducing gateways.
IBM should have a leading position but needs to overcome its consulting services bias and improve edge capabilities along with driving its Watson strategy to a meaningful and functioning capability beyond what the consulting organization can deliver with it.
SAP is trying to get unstuck in many ways. Like Oracle, SAP has both the apps and the analytics capabilities to put into Smart Systems solutions. However, its platform story lags behind Amazon’s and Microsoft’s. SAP, of course, understands that balance sheet issues do matter, and its product suite can be, and has been, used to address these issues in some industries.
Amazon has come out of the backfield and become a significant force in modern information and computing systems. Its maneuvers to create a significant ecosystem of partners coupled with focused “specialist” IoT acquisitions have positioned the company to be a formidable competitor.
In many ways, Microsoft has been the clever player leveraging its legacy strengths from critical horizontal enterprise technologies in infrastructure with its shift to Azure. Coupled with its re-structuring into vertical solution delivery business segments, Microsoft has positioned itself into an enviable position.