How to get usable data to and from cloud infrastructure
Discrete manufacturing needs access to data to ensure the automated systems and machines are operating efficiently and to capacity. However, modern manufacturing or, more pointedly, the data in automation control systems are often siloed. Breaking down the walls to access the data and to make it make sense are just a few of the goals in the world of the Internet of Things (IoT). However, when just having a website may be difficult, how can the machine data be mapped to standard architectures and cloud infrastructures?
Well, that's a long story, but an example is how MachineMetrics in Northampton, Massachusetts, teamed with Amazon Web Services (AWS) to provide a solution that connects the machines and transforms the data. MachineMetrics is an AWS Partner Network Competency partner for the industrial software segment and utilizes advanced AWS services, such as Lambda and Kinesis, to build and scale their architecture. Using AWS technology—the cloud infrastructure—allows MachineMetrics to focus on creating its intellectual property and reproducible designs instead of spending time managing storage and databases and the computer hardware needed to support customers.
However, I've heard stories from both AWS's fans and haters. That had me wondering if AWS is really better than the other guys out there, or is it just interested in dominating the market segment? I asked MachineMetrics' CTO Jacob Lauzier how MachineMetrics can be assured AWS is truly interested in its success.
"AWS’s market segment is cloud infrastructure," says Lauzier. "Very broad, but they see industrial as a huge growth sector, as do we, and they are working with us to provide sales support, marketing resources and joint opportunities, technical guidance and early access to new services. Other cloud platforms, Azure, for example, are getting much better, but AWS still has quite a lead from a technology perspective. I’ve seen a number of articles about how Azure is growing faster—and they will continue to do so—but they’re not necessarily growing at the expense of AWS; they’re growing because the cloud market is growing—and fast."
From a hater’s perspective, I heard about AWS crawling a website, negatively impacting (slowing) the site visitor's experience, so how can the end user be sure that the same thing will not happen on a factory floor system? "I can’t comment on this unless I know the full set of details of the event, but, remember, AWS is a cloud infrastructure company," says Lauzier. "It’s up to its users to write secure applications on that platform. Again, without more context, saying AWS crawled your website is like saying the governor of your state was at fault because someone was speeding. The roads are there as a service. AWS has some policing powers, but there are many details to get it right."
Regardless of whether a customer is using a cloud offering on AWS, Azure or something on-premise, it will need to be built with stringent security standards in place to mitigate risk, continues Lauzier. "MachineMetrics and all cloud implementers are also responsible for providing platforms that can prevent, detect and stop bad actors to maintain a secure, highly available and fault-tolerant system for its users," he says.
Properly implemented, it sounds like AWS products and services will help the machine builders, integrators and end users of the world. "As a cloud infrastructure platform, AWS has the systems in place to allow industrial users to either build their own system or deploy components to their own AWS account that extend platforms like ours through the use of APIs," says Lauzier. "As you know, technology is a series of layers. The more work of others that we can leverage, the faster we can go. Some services offer just tools, and services like ours offer out-of-the-box solutions for quick time-to-value, but then allow those pieces to be extended through the use of APIs."
There are many options for long-term relationships with cloud service providers, but much of it depends on the software the integrators create. "For our customers, it doesn’t matter if we’re built on AWS, Azure, other cloud infrastructure provider platforms or a data center that we manage ourselves," says Lauzier. "The points of contact with our customers are through our user interfaces and APIs. That said, AWS is a great platform, and the support we’re getting through them is allowing us to build on their work so we can focus on the problem we’re solving."
MachineMetrics provides the machine component of the digital factory—IoT for discrete manufacturing. Many manufacturers struggle to use shop-floor data to make the data-driven decisions needed for continuous improvement. MachineMetrics was started to solve this problem. It provides an easy-to-implement platform for connecting to and capturing data from any piece of discrete manufacturing equipment, regardless of brand or age. The platform then feeds this data back to users through dashboards, analytics, notifications and other methods to improve machine performance, productivity and utilization.
