IoT-Platforms: Powering Change – Part 2

Smart Business

IoT-Platforms: Powering Change – Part 2

Data originates all around us and it lives in many places, too. Gathering it and linking it to analytics, storage, and a huge range of specialized apps is a role now being filled by IoT-platforms. From the largest hyperscale cloud provider to narrowly focused niche players, it seems everyone has a slice of the pie.

by Gordon and Eamon McCarthy Earls

IoT-Platforms: Small(er) – maybe beautiful

Let’s start with an analogy. According to Olivier Frank, glob al director for converged servers, edge, and IoT systems at Hewlett Packard Enterprise (HPE) in France, the smartphone has become a pervasive tool in our private and business lives for two key reasons: first, the smartphone converged previously separate functions like music player, camera, computer, and phone into one device; secondly, it’s connected to an app platform that provides countless opportunities to benefit from these converged functions.

Edge systems are converging with operational technology, analytics, and data that used to be separate.

Oliver Frank, HP

IoT-platforms: Oliver Frank - HP

 

“The same thing is happening in the Internet of Things,” Frank says. “[With IoT], edge systems converge previously separate operational technology (OT) functions like data acquisition, industrial networks, and control with standard enterprise IT.” They, in turn, are connected to a rich ecosystem of IoT, analytics, and AI applications to capitalize on the OT data generated in factories, oil rigs, or energy grids, he adds.

Frank, whose company offers its own IoT platform, says these converged edge systems, combined with a platform, are ultimately crucial to achieve the central goal of all industrial IoT initiatives – namely “turning OT data at the edge into intelligence and action to increase efficiency and differentiation.”


/// Read: IoT Platforms Part 1 – SIZE MATTERS: THE BIG FOUR ///


Although the public cloud giants have advantages, in the large and growing universe of IoT platforms there is an enormous variety of smaller, or niche, players that provide offerings which a plethora of pure-play IoT platform providers, such as ThingWorx, Xively, and Ayla, powering various use cases. ThingWorx, which was acquired by PTC in 2013, integrates well with the company’s range of other products that offer a full suite of capabilities for product and service life cycle management (PLM and SLM), and augmented and virtual reality (AR/VR). He also notes that C3.ai offers what it calls a full-stack platform that is, it claims, gaining traction in the market by enabling an analytics platform, rather than leading as a pure-play IoT platform that provides device connectivity and device management.

Next, although not at the scale of the Big Four, HPE and the powerful enterprise application providers – Oracle, SAP, and Salesforce – have also introduced IoT-platforms that are typically based on their own cloud platform. Antonysamy says, “They have seen a fair amount of success tapping into their customer base who already have deployed enterprise apps and have IoT use cases that extend and integrate with these apps.”

In the telecom sector, IoT-platforms from providers such as AT&T, Verizon, Telefonica, and Vodafone have modest market shares but they are in a good position to use their infrastructure and business experience to carve out their own IoT platform territory. Antonysamy notes that their strategies differ from most of the other platform providers because they focus mainly on promoting cellular connectivity and a data pipeline. For example, he notes, one provider has a solution that offers a container-tracking application with the capability to leverage partner devices and their own cellular network and IoT platform.

Industrial IoT (IIoT) platforms represent a fourth sector with services like Schneider Electric EcoStruxure, ABB Ability, and Bosch IoT Suite. “In our experience, the industrial sector tends to compete directly with enterprise IoT-platforms,” says Antonysamy. “What we’re seeing is the industrial solution providers reengineering their own solutions leveraging these platforms and using the platforms as a brand.”

It’s important to make sure that IoT data is enriched with CRM context.

Taksina Eammano, Salesforce

IoT-platforms: Taksina Eammano - Salesforce

 

According to Sabina Kentsch, the marketing and communication spokesperson at Bosch Connected Industry, IoT Suite fulfills this promise by being a PaaS offering based on open standards and open source. She explains that it is intended for IoT software developers, for example, the software portfolio Nexeed, for management of procurement, production, process and logistic data, and is intended specifically for IIoT users. “Nexeed uses services of the Bosch IoT Suite and services of other cloud providers for device management or authentication functionality,” she says. She adds that Nexeed solutions run in the cloud but can also run on-premises as well. Finally, in Antonysamy’s schema, there are the IoT edge providers.

These vendors mainly provide edge and fog (a mini-cloud of edge devices) computing capabilities, complementing cloud-based IoT-platforms by integrating well with them. For example, FogHorn and Cisco are prominent fog providers, while Intel and Digi are leading edge gateway providers. Christian Renaud, a research vice president at 451 Research’s Internet of Things practice, and a former advisor on G20 and European Commission projects, cites recent studies from his firm that underscore the vitality of the platform sector and the strong interest in platforms that are not from the Big Four hyperscale cloud providers. According to the 451 Research report, Voice of the Enterprise – Internet of Things, Vendor Evaluations 2018, of those surveyed, 62 percent use at least two platforms to support their IoT initiatives, while 24 percent use three platform vendors. What is driving this trend toward multiple platform vendors? According to 451, it is the availability of specialist IoT and OT platforms running on cloud infrastructure and able to deliver best of breed capabilities specific to different industries. Some of the multivendor adoption was also attributed to the comparatively early stage of IoT platform adoption with many customers simply experimenting.

Selecting and Implementing

While there are many uses cases drawing organizations toward IoT and toward platform adoptions, it is important to make sure that the IoT data is enriched with customer relationship management (CRM) context for organizations supporting equipment in the field, says Taksina Eammano, VP of Salesforce IoT Go-to Market. That capability can then help field workers to respond to issues in a timely fashion.

“IoT data must be able to connect with customer data so dispatchers can understand which signals coming from connected devices affect what customers and then dispatch the right technician with the right skill set to handle any support issues in the field,” Eammano says.

She notes that organizations need to consider what gaps they currently have in understanding the status of their connected devices. Understanding issues like that can help inform platform choices.

Most platforms provide broad “horizontal” capabilities for IoT and can be utilized to deploy a variety of use cases, according to Antonysamy. He stressed that, at this time, his organization has not seen a strong alignment between specific platforms and use cases. Still, he suggests there are some rules of thumb to consider when choosing a platform:

  • Suitability: Does the customer need a point solution to addresses a specific business unit requirement, or do they want to create a platform that can be leveraged to build multiple solutions?
  • Evaluation of core platform features: This includes device connectivity/management, hot/cold path analytics, along with other capabilities for IoT edge and AI.
  • Total cost of ownership: Including accounting for licenses, subscription, solution development, managed services, etc.
  • Existing cloud platforms in use: This applies to applications as well as data lakes (repositories) and analytics.
  • Greenfeld vs brownfeld: Drawing a comparison between undeveloped land and property that has been developed in the past. Antonysamy suggests considering whether there is an existing solution that must be reengineered with the new platform. Any existing code-base may have to be incorporated.
  • Deployment options: Can the solution be deployed in the cloud or does it have to be on premises?
  • Integration: Do other enterprise applications (either on-premises or in the cloud) have to be taken into account?
  • Non-functional requirements: Scalability, performance, security, and other technical considerations also need to be weighed up.
  • Availability of development accelerators: Time to value can be reduced if solution templates, device agents/plug-and-play devices, and other preconfigured kit can be used.
  • Relationship with the vendor: Ability to influence the roadmap, price negotiation, etc.

The list of rules grows every day and is already far more extensive than the key considerations listed here. The same is true of the number and varieties of platforms and 451 Research says it is already tracking hundreds of IoT-platforms.

It is clear that Antonysamy’s examples form a generalized list of the full complement of commercial IoT platforms available and choosing the best option requires planning and a deep understanding of business requirements.


/// Read: IoT Platforms Part 1 – SIZE MATTERS: THE BIG FOUR ///


Leave a Reply

Your email address will not be published. Required fields are marked *

*