Achieving the Fully Connected Global Operation at Last

Here at Inteva Products, a $3.2 billion global automotive parts manufacturer, we are entering our second decade of cloud computing. However, not absolutely everything we do is in the cloud.

“So after 10 years, you’re still not yet completely cloud-based,” you ask? “What’s been the delay?”

When we began this journey 10 years ago, we experienced a number of network challenges due to physical disruptions that existed around the world. Our global network spans eighteen countries, forty-two locations, and almost 15,000 users. Wireless communications were limited in distance and in bandwidth in remote locations like rural Romania, for example, making cabled networks our only choice. And in many of these places, there was no diverse routing in the last mile.

Massive systems at the perimeter

Our global computing environment was built to accommodate these limitations. Every system that we needed to run  for extended periods of time, with or without a network, was provided on site. Whether it was a manufacturing system to communicate with the shop floor equipment, or some other mission critical system, our teams often built massive, redundant environments at the perimeter of the network. Complex, difficult-to-support systems and processes were the result, with an expense structure that made it difficult to sustain.

But it works and it has been for some time. So, if it ain't broke, why fix it, right? There are two reasons:

1. Governance
2. Cost

Systems management headaches

Managing the distributed computing environment brought several complex issues to the front. First, just keeping the systems running and up to date with software, firmware, and hardware is more than a full-time job for our lean teams. Since local management considers the systems on their floor as part of their operations, working together to maintain upgrades, backups, and other planning can become very fractured. As a result, the amount of work required to develop strong IT governance models for disaster recovery and business planning grows exponentially.

Managing costs in this type of environment is a major headache. With redundant systems, local infrastructure upgrades grow into projects that create a yearly budgeting burden. Additionally, the local teams are often required to become experts in a wide number of difficult technologies.

So, how do you make the change? We do plan on growing our cloud presence and shrinking the on-premise infrastructure as much as possible as we go forward. Our journey hinges on several important opportunities in the computing and communications environment. These include:

• More cost-effective Wide Area Network bandwidth options
• Improved cloud capabilities
• Stronger integration and less hands-on for PC’s at the perimeter

The linchpin to the entire quandary is the Wide Area Network. No longer tied to expensive MPLS networks, current and upcoming options such as SD-WAN, 5G wireless, and other rapidly accelerating technologies offer the chance to dramatically increase bandwidth and significantly reduce cost. Many remote areas have also improved last mile capabilities so that now, our 42-node network has just three locations without diverse cabled paths. The wireless capabilities available now also rival wired environments, and are used in a number of our sites.

The improved offering in computer cycles as a service has also significantly lessened our dependence on local computing. We had the opportunity to leverage Azure as a point solution for a third party product, which also gave us a learning opportunity for further implementations down the road. The goal is to leverage the cloud without requiring completely new skill sets for our systems teams. Virtualization, backup, database, and other options all come with a learning curve that we want to manage.

As companies move to more untethered employees, the internal network now crosses the internet. Having small offices and home offices connected as reliably as any other site on our global network is critical to managing the local devices. For simple processes like changing passwords and receiving company approved patches and updates, the network technology to extend the internal network in a cost-effective manner is here now.

Benefits of centralized computing

So what are the benefits we expect to gain from these network upgrades?

First, we’re enhancing communications between the shop floor and the enterprise like we’ve never been able to in the past. For our company, this means ensuring that operators are trained on equipment before the system will allow them to even login or produce parts. We’re providing a better method for inventory management and visibility into the production environment while walking the shop floor. This can be called the Internet of Things, but to us, it’s just the connected operation.

Second, we’re able to focus shop floor IT teams on supporting the operation and becoming better business analysts. Rather than spending their time monitoring and fixing systems, they’re able to focus their work on exploiting the technology for the benefit of the entire operation. As we move more computing to the core, the local teams won’t need to keep those systems running, because they’ll no longer be on site.

Third, we’re simplifying the governance model. Simplified governance doesn’t sound thrilling until you’ve had to manage a very far flung network with on-premise systems in multiple countries. Do that for ten years and you will definitely be ready to simplify things!