From the August, 2012 Issue of Cabling Installation & Maintenance Magazine
Technology enables high-reliability, zero-latency machine-to-machine applications.
By Daniel Drolet, PCN Technology
Whether for open or closed networks, wireline or wireless communications, machine-to-machine (m2M) technologies and products strive to address the growing needs of m2M network owners worldwide. Meeting these needs requires the owners to pay focused attention to their existing installation base, including functional local automation and control systems already in place. They also must understand the supplemental needs of connecting m2M devices to “the cloud.”
As consumers, we all have been exposed to, and grown accustomed to, digital technologies and how we use them in our daily lives. We all know what an Ethernet jack looks like and that we can connect digital systems easily with Category 5 cabling. We have also assumed that m2M should be easy and because “digital networking seems to be everywhere,” m2M is just the next generation of overall connectivity—the Internet of “things” as opposed to “people.”
We use our iPhones, iPads and other devices to simply download a new Starbucks app while waiting on our venti Americano. I am now finding it even easier to get data from a server in Switzerland than to get the machine data from the receipt printer in front of me. Regardless of whether the machine transaction is made locally or in another hemisphere, these applications all are becoming increasingly intelligent and more distributed.
On a recent flight from Milwaukee, in which “all the seats were first class,” a colleague of mine, David Strumpf, showed me an article from Forbes magazine in which the author had mapped the Internet and connectivity around the world to date. A graphic in the article included many colored lines and graphs wrapping the globe. As Strumpf leaned over from his seat to show me the pictures, he said, “See how far we have to go?” The irony is, just before he made that statement, I thought to myself, “Man, we have come a long way.” We were both right—how fast we have come in the past 20 years, and how far we still have to go.
The world is rapidly moving toward a distributed intelligence, in which m2M will allow machines to interact with the Internet, resulting in benefits that we can only imagine today. This reality is the reason I write this article, and also the reason my colleagues and I have spent the past 12 years focused on the areas of m2M infrastructure and interconnectivity, for the purpose of ensuring critical commercial, industrial, energy, transportation and government systems remain intact while still achieving full benefits of the cloud.
Critical connectivity
Whether the use environment is commercial, energy, healthcare, retail, industrial, transportation, government, military or education, there are billions of devices around the world that require real-time automation and control, and operate with essentially zero latency. These are the systems that keep the trains running on time, literally. They keep our power grids on, manufacturing plants producing, traffic systems in synch, buildings efficient, military mighty, hospitals healing and water flowing. Without these critical m2M networks, many of the world’s functions literally would stop.
These systems cannot fail; they cannot be rebooted nor stand for a software update every other day. They must be available today, tomorrow and decades from now, always available for us as citizens. We have done well to keep our systems secure, reliable and operating in real time. Doing so takes daily management through closed and private systems. But as you read this, a massive digital transition is taking place. We are all familiar with it, and I refer to it as the Industrial Communication Revolution—where closed automation and control systems rapidly are being connected to the Internet/cloud in order to save money, create efficiencies, offer new services, upgrade to the next digital device or better use our mobile devices deeper and farther into buildings. Some refer to this as getting closer to the edge.
As all systems move to the Internet of things, we should ask ourselves the following questions.
1) How will critical automation and control systems, which require essentially zero latency and critical security, be connected to the Internet?
2) Is everyone expected to use wireless exclusively?
3) Can operators of these networks, all around the planet, really be expected to shut down operations, lose revenue, rip-and-replace existing operational wiring for new Category 5 or Category 6, to then connect to a switch in the telecom room?
4) Perhaps the most difficult to solve: How do you get the automation and control engineers (who speak of “screw terminal and DIN rail”) to talk to information technology (IT) engineers (who speak of “7-layer OSI and Ethernet”)—much less have their devices communicate with each other?
These and other questions got us thinking about the current evolution of critical systems, and how they operate in many markets and applications. A product suite that addresses these needs will allow rapid and secure digital m2M transitions so network owners, device manufacturers, integrators and others can achieve the balance of private, real-time critical automation and control systems with low latency, while at the same time being able to access the Internet and its apps, all the way to the edge. Such connectivity will significantly enhance the efficiencies, abilities and features that critical systems could obtain.
Nonetheless, users have justified concerns and fears about connecting critical systems to the Internet, specifically regarding security, in light of some very high-profile hackings. Some hackers are simply waiting to “have fun” with, or do actual and significant harm to, our critical automation systems. However, there are methods to achieve a balance where rapid integration can take place.
So how do we solve such challenges with technology and products that can provide huge benefits and features, not only to network owners around the world, but also to citizens at large? That was a problem/challenge statement that we took on.
Private m2M systems
Private and closed automation m2M systems monitor and control machinery for devices ensuring safety and security, environmental comforts, lighting controls, building systems, public infrastructure and other device-level communications. m2M is not only migrating to the interconnected cloud for remote data management, but also for automated and distributed management of sensors, such as when measurements are taken for temperature, pressure, flow and other automation or control needs. The resulting data is used to make either simple decisions (such as opening or closing a valve), or very complex decisions (such as initiating an emergency shutdown of nuclear power plants or security systems).
m2M systems are deployed to automate and control decision processes for network owners and managers, ensuring that installations, buildings and overall operations are secure, safe, optimal, efficient, productive and perform in real-time without impact from any outside environment. Machines are making these decisions every day. Because of the evolution of the cloud, over time ever-increasing numbers of machines will be connected.
Highly engineered automation and control systems mostly rely on special-purpose, closed network infrastructures for data communication. The protocols used vary and, although standardized, are not always intended for open connectivity to the cloud. This fact poses significant challenges for facilities and installation managers tasked with implementing gateways, extensions or upgrades to the existing network in order to integrate IP-enabled edge devices such as sensors, or to connect the network to the cloud for deploying efficiency tools like Web-based monitoring.
Many of the Supervisory Control And Data Acquisition (SCADA) m2M systems rely on a local area serial network infrastructure for communication. Protocols used include Modbus, Fieldbus, HART, LON, Profibus and numerous others, in which the serial local area network (LAN) architecture is in a multi-drop fashion, with many devices connected to perform critical m2M tasks. As more sensor and SCADA applications need cloud connectivity, these protocols are also being deployed using Category 5 cabling and traditional converters. Regardless of the protocol used, SCADA LANs use some form of encapsulation to communicate on an IP network. For example, Modbus/TCP encapsulates serial data within a TCP packet before sending it across the network.
Driving change
Because of the need for more data and cloud connectivity, several drivers have emerged in recent years that direct transformation within SCADA m2M networks. The major drivers are described here.
Bandwidth—The integration of IP-enabled devices has increased the bandwidth requirements of SCADA LANs significantly. For real-time data acquisition from these sensors, traditional speeds up to 56k are insufficient even for small networks. Larger networks, in fact, call for broadband communication with the ability to handle data rates in excess of 1 Mbit/sec or more. This will only increase.
Legacy support—Most SCADA m2M systems implement mission-critical operations. As a result, facilities managers are heavily constrained by the uptime and performance requirements of the applications implemented on these systems. This implies that integration of new IP-enabled devices or network-upgrade implementations must be conducted without any loss of reliability or functionality of existing operations.
Security—SCADA systems were designed as closed systems, not connected to the cloud. However, with the proliferation of Web-enabled services and the cost-effectiveness of Web-based monitoring, network managers are under tremendous pressure to connect SCADA systems to the cloud using gateways. The implementation of gateways that connect the SCADA network and enable the exchange of data between the LAN and hosted services is straightforward, but security becomes a critical issue because every access point to the SCADA network serves as a potential source of cyber attacks. The efficiency with which Stuxnet disrupted mission-critical operations serves to illustrate how pervasive the effects could be in such an event. Security-at-the-edge has therefore become a key consideration in the design of SCADA m2M upgrades and migrations.
Until recently, traditional thought was that network and resulting infrastructure upgrades could only be completed by implementing one of few technology solutions, independently or connected to each other.
One such solution is wireless, where the use of standard off-the-shelf wireless technologies for LAN connectivity could pose challenges in mission-critical applications. Connectivity and latency issues could come into play when wireless systems must operate in critical SCADA environments, where radio-frequency interference (RFI) or electromagnetic interference (EMI) and electrical noise are high. Also, dedicated bandwidth above 1 Mbit/sec must be available consistently. And maintaining control over all access points to support security requirements presents a complex challenge.
As a result, while wireless enjoys significant advantages in its simplicity of deployment, it may not always be a preferred choice for the only secure broadband SCADA LAN infrastructure solution.
The second option is to entertain the deployment of a new structured cabling infrastructure and IP networking equipment within the SCADA m2M system to integrate new IP-enabled devices and connect SCADA LAN to a gateway in the premises.
This approach ultimately results in two separate networks between the legacy serial SCADA operations and the new IP infrastructure. Additionally, any architecture redesign may involve the use of serial-to-IP converters and the implementation of the legacy operations on the new structured cabling would substantially increase the technical risks in maintaining uptime and performance of the legacy applications. So while new structured cabling itself may satisfy the requirements for the integration of IP-enabled devices, it does involve a significant capital budget, technical risk and integration issues between legacy operations and the new infrastructure.
A new option
Since 2010, networking technology has been available to enable the rapid deployment of Ethernet extension, IP gateways and IP-enabled networks by using existing functioning serial infrastructure (e.g. RS-485, RS-422, RS-232 and others) while preserving the serial data that is already there. The technology uses the existing multi-drop network architecture. The products equipped with this technology include network routers, switches, extenders and serial network stabilizers (Photography of commercially available products is on page 24).
In the context of SCADA m2M LANs, the available technology enables the deployment of secure, broadband Ethernet/IP networks on existing, functional wiring without impacting any of the existing SCADA operations that must be maintained, such as Modbus communication between SCADA remote terminal units and programmable logic controllers.
Network owners can manage their migration to the cloud by instantly mixing and matching where and how they choose. Whereas the rip-and-replace approach requires system shutdown, this new technology enables the deployment of new services, efficiencies and applications without shutting down critical automation and control networks. Through this technology, an Ethernet LAN can be dropped into and deployed within a SCADA m2M system in order to enable the deployment of new, multi-drop IP devices in order to connect the SCADA system to a gateway, and enable the transport of compliant IP security standards. Meanwhile, legacy serial operations and devices are left alone.
The ability to have a multi-drop Ethernet environment wherever there is existing copper cabling also provides rapid cloud connectivity for the deployment of new IP-based services and solutions within the context of a serial network architecture. The combination of this network technology and standard wireless connectivity for the wide area network enables network owners to move m2M data farther into infrastructures.
The most prominent examples of this technology in use are the Wayne Connect IP-485 offering, deployed at fueling stations in the United States and Canada. The system is used at these fueling stations to ensure that serial networks do not become bogged down, so automation and control can take place simultaneously with IP and Ethernet upgrades for PCI, digital media, remote connectivity and other digital applications.
The Wayne Connect system enables the integration of Wayne equipment to the Wayne network operations center for the delivery of Web-enabled services. The system has demonstrated dedicated IP network bandwidth of 1 to 4 Mbits/sec per edge switch for Ethernet traffic, while all existing legacy serial functionality is maintained with jitter-free, zero-latency transport for RS-485 bidirectional network traffic.
The technology is also being deployed in building automation, control, security, fire and safety networks. Products are used to upgrade existing serial networking infrastructure without impacting functionality, performance or reliability of the existing legacy operations. It allows building systems to be maintained while a managed migration is performed, enabling building owners to inject and interweave Ethernet anywhere they want within a serial network without opening up their entire infrastructure to the Internet or an intranet.
Relatively new technology provides a new approach to ubiquitous IP LAN connectivity, without the need for construction and rewiring in SCADA systems. What traditionally took months of planning and weeks of down time can be completed in hours. Its capabilities can provide facilities managers with peace of mind about the costs and risks involved in LAN upgrades, while also making cloud connectivity all the way to the edge more readily accessible and simpler to install. ::
Daniel Drolet is executive vice president at PCN Technology Inc. www.pcntechnology.com). The Wayne Connect IP-485 system is a private-label offering of PCN Technology’s IP-485 technology. Also contributing to this article were PCN Technology’s president and CEO Venkat Shastri, CTO David Strumpf and director of marketing Ray Kahue.