This is Part 9 of our “25 Things to Consider when Choosing an OPC Tunnel” series. Will a tunneler support redundant tunnel paths?
Redundancy is typically a more complex topic than it might sound especially if your automation control architecture is extensive and has multiple layers. Redundancy has different meanings to different people for different applications.
This blog post will outline reasons why it's important for an OPC tunneling solution to support redundant paths for applications where it is essential minimize the risks of failure when transferring your control data and what questions to ask.
Since, in the real world, the term redundancy can have many different meanings depending on your specific application, it's important to first determine which components in your system you are trying to protect from the failure. Do you need redundant PLC’s, redundant OPC Servers, redundant tunnels and redundant OPC Clients?
Since there can be multiple decisions to be made, when you start talking about redundancy it is critical that the architecture be clearly thought out before choosing a software solution or solutions. This is because it is not necessarily a given that automation control software will support redundancy on any or all levels.
This is just as true of any OPC tunneling solutions you are either considering or currently using as it is of any other automation control software such as OPC servers or the actual HMI or SCADA system in your architecture. It's crucial to discuss redundancy support with the supplier about available redundancy options and performance of the solution's failover functionality. Consider the following questions when talking to the potential supplier:
- Does the solution even support redundancy/failover?
If a tunneler doesn't even support redundancy on any level, you have to consider the cost of an unavailable connection in your system, even if it's for only a few seconds. Weigh that cost against the peace of mind provided by a tunneling solution that does support redundancy.
- How long does the failover from the primary tunnel connection to the secondary connection actually take? (less than a second, more?)
Pardon the phrase, but "time is money" certainly applies in the realm of automation control. The longer failover in a redundant system takes, the longer you're potentially down and losing money in the process. Make sure your tunneler isn't wasting precious uptime and is failing over as quickly as possible.
- On what level does the tunneler support redundancy? (Just for the tunnel, for the OPC Server, both?)
A system is only as strong as its weakest link, as they say. This applies to redundant systems, as well - if only one layer of your system's architecture is redundant, you're still vulnerable if any of the non-redundant components should fail. Determining your risk at all levels of the architecture and preparing with appropriate redundancy measures should always be part of the decision making process when choosing such software solutions.
Reputable suppliers will have multiple options for redundancy and associated switchover times and will discuss pros and cons of each method, as well as being able to discuss performance and behavior during a failover.
Before purchasing a tunneler, make sure you've determined your application's redundancy needs. If your data transfer and connection uptime is critical, make sure you ask the right questions to ensure the tunneling solution you choose will adequately mitigate the risks of downtime and loss of productivity. To reiterate from our other posts, choosing an effective tunneler that takes into account your application requirements will make a big difference in your operational effectiveness, resiliency, and profitability, as well as minimizing downtime. Learn about the other reasons in the free whitepaper “25 Considerations when choosing a tunneling solution”.
