No-Nonsense Networking

Machine Builders Must Expand Their Awareness Beyond Their Core Components to Embrace a Wider Circle of Related Equipment, Applications and Business Goals, and the Responsibility for Them

By Jim Montague

Share Print Related RSS

Practically the first time I interviewed somebody and wrote a story about it, I realized that I was just a pipeline from them to whoever was going to receive their information. Consequently, being sort of a "network switch" myself, I know the pathway isn't as important as the content, but we do need some assistance to deliver our data.

And, just like the sensors and controls they serve, networks and their users also need to be conscious of the overall context in which they function. I was reminded of this by several sources in this issue's cover article, "Monitor and Mend," on network health. They jointly agree that everyone from designers to operators must expand their awareness beyond their core components to embrace a wider circle of related equipment, applications and business goals, and the responsibility for them.

As with any control and automation project, the first step in network health is to account for and audit all existing components and systems, evaluate if their capabilities meet present and future business goals, and then decide what new functions and capacity need to be added. In most of today's industrial networks, basic health means using managed Ethernet switches and network management software (NMS), but there are many other common-sense best practices that can help, too.

"Simply reading installation manuals can prevent a lot of problems," says Patrik Boo, global product manager with the Process Automation (PA) service group at ABB. "Many difficulties come from users not fully understanding how to apply and manage certain devices, and assuming facts that aren't true. This is especially true for relatively new technologies such as NMS and managed switches. For example, it's important to know that Ethernet is incredibly robust and can operate on the edge of a failure for years, so users need to know what their network alerts mean to address these issues."

After inventorying and documenting their present systems, users often must decide if they have enough data capacity, or if they need switches and cables with more bandwidth and throughput, or if they should divide their network into several subnetworks. Next, they need to choose what new design or topology they might need, what type of NMS is required, and the right combination of point-to-point, fieldbuses, Ethernet and wireless equipment will serve them best.

Boo reports that the 800xA license option PC, Network and Software Monitoring (PNSM) uses SNMP/SNMP Trap to supervise the IT infrastructure in an 800xA system configuration. The SNMP protocol reads diagnostic data from typical server/client/network equipment, and the data is published as OPC properties in the 800xA system. This OPC data is typically written by Asset Monitors to generate automatic condition monitoring alarms from the IT equipment.

In fact, even though some networks or subnetworks now exist virtually on rack-mounted servers, they too can be organized into distinct, software-based sections to help improve security. "For instance, 800xA runs in a virtual environment, which allows the system to operate more efficiently and in some case more securely," Boo says. "Whether virtual or physical, networks still must be organized logically and pragmatically. I think 80–90% is just making sure they're designed and installed correctly from the beginning. So, when you go to verify network throughput later and some issue comes up, you'll know where to go and what to check when you get there."

Boo adds, "We also have a Cyber Security Fingerprint service that includes entry-level security checks. We collect data from our customers' sites, assess what network security methods they're presently employing, compare them to best practices, identify their individual strengths and weaknesses, and write a report for them with a prioritized action plan. This can include establishing a policy for deploying software patches from Microsoft or providing staff with more security training."

These are all great tools for improving network health, but they're only helpful if they actually get used. Unfortunately, one of the biggest obstacles to employing them is that many engineers and other professionals are very reluctant to reach out and ask for help — like all those dads who never wanted to ask for directions. I was always concerned that asking a lot of questions might make me look stupid — until it became my job and I knew for sure, but got paid for it. Likewise, technology is changing so fast these days that it's essential to request and get the assistance we need, even if it's uncomfortable to do so.

Share Print Reprints Permissions

What are your comments?

Join the discussion today. Login Here.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments