Bridging IT and OT

Wednesday, November 11, 2015 @ 03:11 PM gHale

EDITOR’S NOTE: This column emanates from a white paper entitled “The IT/OT Convergence — Bridging the Gap.”
By Frank Williams
Operational technology (OT) networks are progressing to the point where a transformation is in full swing.

Sometimes referred to as the Industrial Internet of Things (IIoT), Industry 4.0, or Digitization, it is clear industrial firms have begun to use available technologies to completely revamp their business model. This realization by industrial professionals is key in moving the IT/OT convergence to reality.

IT Getting an OT Education
Stuxnet Loaded by Iran Double Agents
Blackhat: Hacking a Chemical Plant
Blackhat: Recovering from Shamoon

IIoT is a sensor-focused network designed to completely integrate into control-focused networks at the plant level and data-focused networks at the enterprise level.

Because the goal of IIoT is to maximize the amount of data throughput to the enterprise and the control system, IIoT sensor networks connect to the cloud (for asset management, simulation and modeling, and similar non real-time tasks) and to the control system for real-time bidirectional control, and with enterprise requirements served by the cloud. This means that each data point may have two or more destinations in the network.

We should expect the complexity of IIoT networks, with virtualization and cloud computing, along with hardware and software as services to become greater. With that complexity, it means the need to continuously monitor and run diagnostics on these networks will continue to ratchet upward.

Just how complex will IIoT networks become? Current numbers of sensors at typical process plants cluster around 40,000 sensors. The IIoT will increase those numbers exponentially to something over 250,000 sensors per plant. Each of those sensors will be producing near real-time data at an update rate of four times a minute, or 250 milliseconds per datum. That means each sensor will be producing over 5,000 data points per day. That’s 1.44 billion data points per plant, per day. Each of those sensors needs to end up monitored and diagnostically checked for proper operation as part of the network.

It is essential the IT/OT network administrator have tools that allow true and total visibility of the network in real-time, permitting the network administrator to view, review, and perform analyses on the data at any point. Diagnostics and alerts are capable of telling the administrator when a sensor or a sensor network goes down, so personnel can perform emergency service.

Converged Networks
In the future, every enterprise network will be made up of sub-networks that are either data-centered, user-centered, or sensor-centered. Each of these networks will interface directly with all other networks in the enterprise.

This will make the job of the network administrator significantly more difficult than it is today, with mostly independent networks in the enterprise or on the plant floor. These networks will need to come together to provide the benefits that Big Data, analytics, the Internet of Things, and the cloud will provide.

When planning for the future, look for network tools that will enable you to poll the entire network in less than 60 seconds. No longer will you be able to only look at ‘mission-critical’ devices because your network infrastructure monitor can only poll a network every 5 minutes or so. You must know what all devices are doing — or not doing — on your network, all the time. Fast polling will be a mandatory feature. And the growth of networks will demand the standard be less than 60 second polling for your entire network.

Scalability will be another key factor. Network solutions that easily scale to meet the explosive growth rate of your network, and do so, in a CAPEX/OPEX friendly way is crucial.

And finally, it will no longer be acceptable to average the data; granularity of data for quick network forensics will be the norm. Your network monitoring solution must be able to provide granularity to all of your data.

Automation industry veteran Frank Williams is the chief executive at Statseeker, a provider of network monitoring technology. For more details click here to view the white paper entitled “The IT/OT Convergence – Bridging the Gap”.