Site icon IT World Canada

9 technologies to supercharge your Internet of Things deployment

Shutterstock

Ever wonder what technology components underlie the Internet of Things (IoT)? As the explosion of IoT continues, understanding the technology components of IoT becomes important to business success. Here’s a quick overview that should help you make more informed technology choices when implementing IoT for business value.

IoT devices, typically sensors, are continuing to decrease in size, electrical power consumption and cost. At the same time IoT devices are increasing in telecommunications and data processing capability. For examples of consumer IoT devices, look at the Nest Labs website. In the future, we can expect IoT devices to cooperate more, even across companies, as store-and-forward communication devices. This capability will control today’s significant data gathering cost.

These IoT device advances will make ever more ambitious IoT applications possible.

Networks

Local Area Networks (LAN) are becoming faster, more resilient and sophisticated in their management features. Wi-Fi is becoming more ubiquitous. The available technologies for Wide Area Networks (WAN) are increasing. The cellular telephone network will soon be upgraded to 5G for faster speeds and more capacity. More and more fiber optic cables are being strung across the planet. Satellite communication is becoming easier to use and more cost-effective.

These network technology advances make it possible to collect the detailed data from the rapidly increasing number of IoT devices scattered across the planet at a cost-effective price.

Data analytics

IoT data is labeled Big Data because it is so voluminous. The huge volume makes IoT data useless without data analytics software.

Leading data analytics software development tools include IBM Watson, Microsoft Power BI, Qlik Sense, SAS Viya, Tableau, and Tibco Spotfire.
In the future, expect data analytics to emphasize data exploration and expect data exploration to become more autonomous as Artificial Intelligence (AI) is introduced into data analytics software.

Computing power

General purpose processor chips are increasing in computing power without increasing in price. These chips provide plenty of power to process the torrents of IoT data rushing into data centers.
Older processor chips are falling in price and offer more than enough processing power for many IoT applications that collect, crunch and present data. These computing power advances mean that ever more ambitious IoT applications are becoming possible.

Co-processors

Field-Programmable Gate Arrays (FPGAs) and Graphics Processing Units (GPUs) provide high-performance to computationally intensive systems. Co-processors offer higher performance per Watt for specialized instruction sets compared to general-purpose processor chips. A good example is the specialized chips Google designed and had built for voice recognition on Android smartphones and personal assistants.

These co-processors significantly reduce the elapsed time and cost required to process and summarize the large volumes of time-series data that IoT devices typically generate.

Memory

The cost of memory in computing is continuing to decrease. The low price makes it possible to add at least some memory to even the smallest, cheapest IoT devices to retain more data.
Low-cost memory enables a store-and-forward approach to communicating the IoT data to a central data ccentre for storage and processing. Store-and-forward reduces the need for a more expensive, highly available network required to gather the IoT data when IoT devices have no memory.

These memory cost reductions mean that ever more capable IoT devices can be deployed at decreasing cost.

Storage

The cost of disk and tape storage is continuing to decrease. The low price makes it possible to keep the large volume of detailed data that IoT devices generate indefinitely.

Some companies are summarizing IoT data to contain the growth of storage. Others are recognizing that IoT data is emerging as a strategic dataset that confers competitive advantage to its owner. These storage cost reductions mean that companies can err on the side of caution and keep as much IoT data as possible.

Database management systems

Advanced database management systems (DBMS) are required to manage the large volume of detailed data that IoT devices typically generate.
Examples of large DBMS’s include:

  1. SQL: IBM DB2, Microsoft SQL Server, Oracle Database.
  2. NoSQL: Apache Spark, MongoDB, Apache Cassandra.

These DBMS advances make it possible to manage and retrieve large volume of IoT data at high speed and at an acceptable price.

Application development tools

To derive value from IoT data requires an IoT application that analyzes and presents the data in a form that is useful for decision-making. IoT applications are typically custom applications that are developed using application development tools. Examples of leading application development tools that are suited for the job include R and Python.

Cloud services

If thinking about all these technology components associated with IoT is giving you a headache, you can outsource large parts of your IoT initiative to a cloud service provider. The major players are Amazon AWS, Google Cloud Platform, IBM Cloud, and Microsoft Azure.

The major advantages of outsourcing include:

  1. Little or no capital required for the computing infrastructure.
  2. Outsourcing all the operation work, including updating and patching the computing infrastructure, that is a major distraction for many organizations.
  3. Access to advanced data management and AI software functionality that would be prohibitively costly to build and maintain for one organization.

These cloud service advances make it possible to reduce the time to IoT value. Shorter time to value can mean the difference between a breakthrough product or an also-ran offering.

What technology approach have you found useful to ensure that your IoT initiative delivers business value? Let us know in the comments below.

Exit mobile version