Module 6

Scalability of IoT Solutions

 Roadmap for developing complete IoT solutions, Strategies for implementation, key Milestone,

 Scalability of IoT Solutions, Methods, platforms and tools.

 Web and Mobile Interfaces

NASSCOM
 Design roadmap for developing the complete IoT solution

 Define different stages for implementing IoT solutions with key milestones to be achieved

 Define how the IoT solution will scale from a proof of concept to an actual solution with thousands of devices and

real production constrains

 Discuss methods, platforms and tools to be adopted to scale IoT solution

 Design web and mobile interfaces that can handle multiple devices at scale
Design roadmap for developing the complete IoT solution

Product/Solution Roadmap
- a plan for executing IoT Solution
- high-level visual summary that maps out the vision and direction of the product offering over time

Factors to be considered while designing Solution roadmap

Visions, solution offerings, devices, Costs, duration, internal stakeholders

Product Roadmap - Goals:

Describe the vision and strategy
Provide a guiding document for executing the strategy
Get internal stakeholders in alignment
Facilitate discussion of options and scenario planning
Help communicate with external stakeholders, including customers
IoT Solution Roadmap Development
IoT product is a combination of different technologies working together at different layers of IoT technology stack. In order to
have a to-the-point IoT product roadmap, the layers of the stack must be mapped to the task for which the solution is
developed.

Example - Water Pump

IoT Solution Objectives
 Vibration Monitoring
 Temperature Monitoring Data Monitoring
 Electric usage Monitoring
 Maintenance Service
Analytics
 Emergency Shutdown

IoT Technology Stack

 Device Hardware and Software – Device Layer
 Communication Protocols – Network Layer
 Cloud Platform
Application Layer
 Application Programming Interface
 IoT Solution Roadmap Development - Water Pump Use Case

IoT Layer Device Network Application

IoT Technology Device Device Communication Application

Cloud Platform
Stack Hardware Software protocols Programming interface

Read the Transfer the data Display the data on the

Vibration Monitoring Add Sensor Store the data
sensor value to cloud dashboard

Maintenance Predictive Display the data on the

Service Analytics dashboard

Display the data on the

Emergency Shutdown the Real-time Transfer the data Generate an
dashboard. Send alert
Shutdown pump Analytics to cloud alert
the user.
IoT Solution Objectives
Considerations for developing IoT Solutions
1. Proven Architecture
Leverage a reference architecture that already exists, and tune it for the desired domain and solution.
IoT solutions typically align to certain patterns that are well-proven. These patterns align to the following aspects of IoT
solutions:
 Patterns on the edge: communication, Analytics, management, security, intelligence
 Patterns for ingestion: scale
Data Ingestion – data gathering, transferring to the processing systems where the data is stored, analyzed and accessed.
 Patterns for processing: extensibility
 Patterns for actions: communication, integration

2. The Right Topology and Protocol

Factors need to be considered when choosing the topology and the protocol:
How many nodes are to be connected?
What is the range expected (Example: to cover a factory of a few square km, or a home of a few square meters)?
What constraints do the nodes have (Example: battery and temperature)?
What already exists in the landscape (Example: existing Wi-Fi connectivity)?
What are the existing security requirements?
Protocols

3. Middleware Platform Choices

On the middleware side ingestion, processing, storage and integration is housed
4. Building for Scale
Scalability for ingestion, processing, and storage

5. Security
Different stages for implementing IoT solutions
 Requirements and Functionality
 Brining on Partners
 PoC, Prototyping, Piloting
 Testing IoT solution implementation for the first time
 Refining and Solution Development
 Scaling
 Lifecycle Management

 Device connectivity and data forwarding

 Real-time monitoring
 Data Analytics Adding use case to an existing IoT solution
(Considered while creating PoC, Prototyping and Piloting)
 Action Plan
 Enhancing onboard intelligence
How the IoT solution will scale from a proof of concept to an actual solution?
Internet of Things
- IoT enables connected operations and connected products
Advantages of Internet of Things
 Increased efficiency and productivity
 Reduced cost
 Enhanced safety
 Security
 Reduced risk
 Open new revenue streams
 Better usage of resources

Though the IoT implementation objectives are clear, a recent survey found that only 25% of IoT
projects had moved from the proof-of-concept phase to the use phase.

Why organizations fails to scale their projects beyond proof of concept?

During proof of concept, organizations focus on proving the technology works, at the expense of
taking a longer view to how they will operationalize and commercialize that technology at scale.
Three dimensions of scale
 The operational: Does the organization have the resources to operate a large-scale IoT implementation or adopt
new business processes that the new solution demands?
For example, updating devices in the field as needed or deploying solutions at thousands of sites globally.
 The economic: Does the organization have the ability to scale the solution in a way that maximizes the business
value and ensures the highest return on investment (ROI)?
 The technical: Does the organization have the technology foundation to support implementation at scale? For
example, connecting millions of devices in the field and managing the enormous amount of data they collect.
Operational scaling
Transforming business processes
When moving beyond proof of concept, the real value of IoT comes from using solutions to modify existing business processes and
create new processes.
Effectively incorporating IoT into business processes typically requires acknowledging and documenting changes to existing business
processes and new processes and then training the developers/end users. Developers/end users need to understand how to read the
new data, whether it’s provided in a dashboard or another format, and be able to act on that data.
For example, in case of condition monitoring, the operations team needs to be able to use the IoT data to understand when equipment
has failed or is at risk and use that to inform how they’re allocating maintenance resources. It is necessary to ensure that
developers/designers understand not only the technical aspects of the solution, but also more importantly the changes in process that
are necessary to realize the full benefits.
Managing operations in the field
The operational dimension of scale also comes into play after the IoT solution is deployed. Every aspect of the IoT solution needs to be
maintained, including hardware, edge components, connectivity, and cloud.
But because IoT involves physical hardware in the field, there are elements that require a human touch. For example, sensors or
gateways may need to be replaced. Organizations need to determine who is responsible for replacing these components, whether that
be IT or OT. They also need to think about whether they have the ability internally to roll a truck to the site to perform maintenance or
replacement or whether it’s better to outsource some of this work to external contractors.
Economic scaling
Organizations implement IoT at scale because of the value it will bring to the business, whether that be cost savings or new revenue
streams. with the right approach, organizations can maximize ROI and contain TCO, reaching their economic objectives

Delivering return on investment

organizations can expedite understanding of the potential ROI and value of an IoT project by performing a detailed assessment of the
operational environment, which can more accurately identify the potential areas of value, help guide the direction of the proof of
concept, and build a more accurate picture of ROI.
Organizations should use their proof of concept to not only prove the technology works, but also demonstrate ROI.

Managing total cost of ownership

TCO primarily comprises the cost of devices, connectivity, and compute infrastructure. When taking IoT from proof of concept to scale,
it’s essential to have controlled, predictable costs for these elements. . It’s essential to understand the business case for connecting
different devices and prioritize which devices to connect based on what will drive the most value. Understanding the per-facility or per-
device price enables organizations to make strategic scaling decisions: they can better predict ROI and choose how and when to scale.
Technical scaling
Two of the key challenges related to scale are connecting and managing devices and managing data.

Connecting and managing IoT devices at scale

Organizations need to plan for the thousands, sometimes millions, of devices required to scale their IoT solutions. During proof of
concept, improvised, manual approaches to connecting and managing devices may show the technology is viable, but scaling manual
processes won’t cut it when dealing with thousands of devices
Identifying and onboarding devices
The first step is to establish the connection to the cloud and configure the device to perform its intended task. In a proof of concept with
only dozens of devices, it’s not hard to onboard devices manually, but at scale it’s impracticable. Organizations should take a zero-touch
provisioning approach to enable automated provisioning at scale. With zero-touch provisioning, a handshake is triggered when a device
is powered on which initiates onboarding and a subsequent automated provisioning process. Because the initial handshake includes
authentication of identity, credentialing of trusted certificates, and security of the device and data, zero-touch provisioning not only
eases the onboarding of devices, but also supports a strong security posture. To keep IoT solutions running as intended, it’s essential to
ensure that they are able to connect reliably to the cloud.
Managing connectivity
A best practice for connecting thousands or millions of devices across the globe is to use multiple instances of a core connectivity and
device management service like Azure IoT Hub
Updating and maintaining devices
Managing IoT data at scale
data is only valuable if you’re able to use it, which means ingesting it, transforming it, and applying rules for data processing
Ingesting data
Collecting data from different sources and transferring to the processing system.
Transforming and enriching data
In the majority of IoT implementations, the devices involved will be heterogeneous and will likely send data in different formats. A best
practice is to use a common data model which provides a shared language for all your devices and solutions, considerably reducing
effort for solution developers. the data coming from devices will be raw, so it can’t be used with your IoT solutions without further
transformation. To address this, organizations should enrich their IoT data with contextual or reference data for further processing.
Applying a rules engine
Users within organizations need to be able to define rules for data processing. For instance, some data needs to be processed
immediately to enable real-time use cases (hot path), whereas other data can be sent to storage to be queried later in larger batches
(cold path).
Establishing an enterprise IoT platform