The Internet of Things – Four Things CIOs and IT Can Do Today
Today’s blog is for information technology leaders who know they want to explore the Internet of Things (IoT) as a powerful capability and want to learn how they can get started now building a technical foundation for it. You’ve probably heard a lot of discussion about the Internet of Things (IoT) over the past few years. It’s here now, in its earliest forms, and it’s generating industry momentum quickly. Part of the reason is that IoT is a perfect third partner for mobile and cloud computing. And over time, it will be the genesis for the IT function to enable new business models and revenue growth opportunities.
The Internet of Things (IoT) is in the early stages of the “S” curve and will certainly continue to evolve. We will see new business models and technology solutions take shape, leveraging data harvested from smart and connected devices that will permeate our work and personal lives.
Already, companies are embracing IoT supported strategies that connect customers to their devices, their devices to each other and all of those back to the enterprise.
To address the IoT opportunity, we will first start out with some definitions and then discuss some practical challenges CIOs should as they consider as they prepare for the deployment of IoT within their enterprise.
The Internet of Things – What Is It?
Ask two people and you’ll get three answers on what the Internet of Things actually is. Even companies like Cisco, IBM and SAP differ in their definition, with Cisco branding it as the Internet of Everything (IoE). IBM suggests it will be a completely separate web, based partially on the protocols that make our current worldwide web work, but requiring new standards for device control, communication and data security. General Electric calls it the Industrial Internet.
For our purposes, we’re going to assume IoT means bringing a whole range of physical objects online—from complex machines like automobiles and mining equipment to simpler objects such as home appliances and thermostats.
The worldwide web we all use today focuses on people and servers. The IoT web, on the other hand, will be optimized for machine-to-machine (M2M) communication. It will need to be efficient at bringing inanimate objects to life online through embedded hardware and low-overhead data collection and integration services.
For example, embedded sensors in products use mobile technology to connect to each other and to cloud services and hybrid data centers, all the while feeding Big Data to analytics applications. This ecosystem is enabling products to communicate their condition and usage to users as well as the manufacturer. The possibilities to harness this information and use it to improve the performance and value of IoT enabled devices are nearly infinite.
However, there is much uncertainty in not knowing all of the ways in which IoT will technicallymanifest itself. It’s widely acknowledged that IoT has potential security challenges that will require new standards and protocols that may not exist yet. But that hasn’t stopped companies from enabling solutions based on today’s IoT capabilities.
Almost 20 billion new products will connect to the Internet of Things over the next five years.
Four Ways to Prepare for IoT
1. Develop an IoT Strategy
IoT will enhance existing products and services, enable new lines of business and transform the ways existing businesses operate, from creating new sales channels to improving supply chain efficiency.
New business models will grow as IoT capabilities continue to mature, and the implications are enormous for IT, CIOs and the enterprise. CIOs and senior IT leaders will find plenty of challenge in collaborating with their business counterparts to exploit the capabilities of IoT within their enterprise, but the impact on the business can be significant.
How Can I Get Started?
The IoT is ideally suited for products where the value of the product can be enhanced by knowing more about certain parameters associated with the product – its usage or utilization, its performance, its state or condition and its location over a certain time period. The essence of IoT today is capturing these parameters in real-time or near real-time from smart and connected devices. As these devices can spin off quite a bit of data, the challenge is in analyzing these parameters to identify trends and other useful business insights. These insights can then be used to 1) improve the customer value of the product directly (e.g., improve performance or reliability) or 2) create ancillary services based on the product’s usage (e.g., provide suggestions on how the product can improve customer benefits).
CIOs and senior IT leaders should engage business leaders in a review of current product and service offerings to determine if any are candidates for the development of an IoT enabled solution. The best approach is to start small (i.e., avoid a “big bang” approach) with the goal of learning more about opportunities to enhance products (using the parameters mentioned above), understanding complementary IoT technologies (third party vendors) and the size of the market for smart and connected devices (customer receptivity). As an IoT strategy begins to emerge, CIOs will then need to consider the implications on the enterprise computing environment.
2. Plan for New Demands on Infrastructure and Operations
As said earlier, 20 billion devices will be online by 2020. Let’s combine that with another interesting statistic. Most IT professionals, some 57%, report that their current networks are already at capacity. Given that 46% of IT professionals also believe that IoT devices will be coming online and talking to existing networks, this is a recipe for disaster.
For those running private clouds, the time is ripe for additional investment in IT infrastructure. For those who need to scale up quickly, cloud solutions provide immediate scalability. Plus, the proven capabilities of Amazon, Microsoft and others to ensure that when the data storage demand hits, you’re already covered.
How Can I Get Started?
Consider the operational management of all of this. You’ll need to deploy, track, reassign and de-provision all manner of sensors and devices lined up at the edge of your network, all chatting and requesting valuable resources to service themselves, customers and other devices. So, create or augment your asset management service and Configuration Management Database (CMDB) so your IoT deployment doesn’t resemble stadiums full of strangers all shouting at the same time.
3. Embrace Big Data
The initial barrier for most IoT implementations is the infrastructure required to get devices talking to the network. Then, as IoT data is generated, it needs to be harvested, stored and processed for analytics. This is the domain of Big Data, and it will determine much of the effectiveness of your IoT strategy.
Most private data centers are geared towards scaling in terms of terabytes. When scaling in units of petabytes and exabytes becomes the new reality, this may not mean simply scaling up or out in ways that the enterprise hadn’t factored, it may mean completely different approaches and wholesale replacement of existing solutions. Few IT shops today are ready for this amount of data processing and analytics.
How Can I Get Started?
Big Data is another area where you may want to push the solution outside your physical walls and adopt one or more cloud solutions to provide the scalability and sheer processing power required to handle the demand. Amazon RedShift, Hortonworks, Cloudera, Mongo and other providers are already scaling at this level, and have invested time and money into data solutions designed with the intent of supporting not just large volumes of data, but IoT specifically.
4. Update Your Services-Oriented Infrastructure
So now that you have a network or cloud solution that will handle the traffic, storage and analytics, you need to integrate all of this at the application layer. Depending on the application of the device, you will have latency requirements varying from sub-second (traffic control systems) to near real-time (banking and financial systems). Some messages will also directly request enterprise services and workflows that exist in ordering systems and ERP systems.
Sensors and smart devices of all sorts will become smaller and smaller with minimal amounts of local processing power required. The ability to process large payloads of SOA-generated XML, an expensive operation required of many Simple Object Access Protocol (SOAP) clients, may not be desirable or possible. However, RESTful services provide a far more lightweight method of dealing with data payloads.
How Can I Get Started?
Prepare a REST services capability either in place of, or in front of, your existing SOAP-based SOA infrastructure to readily communicate information from your IoT world to your bus. Having fast, secure services accessible from the edge of your network will be a requirement, and REST services will help.
Imagine being able to embed intelligence in your products and monitor their entire lifecycle, from initial purchase to obsolescence. Consider the value of applying proactive diagnostics to equipment all over the manufacturing floor to fix issues before they become problems. Think of the advantages of making products that communicate with each other in the field. And if you see robotics playing a role in your future, the Internet of Things will become central to your plans.
Watch what happens as IoT transforms the world with advancements we never thought possible.