Understanding the Meta, Comcast, and LinkedIn Outages


Enable Greater Automation with Greater Visibility

By Ian Waters
| | 9 min read


Industrial companies are pursuing automation to save money and achieve efficiencies. They will require a host of systems working in unison to succeed.

Watching organizations press forward in their quest to finally realize the promise of IoT and automation is nothing new. Large industrial companies—in mining, manufacturing, aerospace, and other fields—are pursuing automation to streamline staffing, improve worker safety, achieve greater operational efficiencies, and, of course, save money.

The trend has been steadily unfolding over the past five to eight years. But there have been bumps along the way.

Systems Working in Perfect Unison

For automation to truly succeed, it requires a host of different systems working together in perfect unison. It also depends heavily on the global Internet. But as we all know, disparate systems don’t always communicate effectively, and Internet outages are more common than we’d like to admit. Getting multiple systems in multiple environments to function properly with optimal connectivity and limited latency is a monumental challenge. And because the Internet is a web of shared, best-effort networks, its performance is nearly impossible to predict. 

The complexity of automation is equally intimidating. In big facilities, a huge volume of data is generated at the edge. That data needs to be recorded, stored, and analyzed with relevant information, then sent back across the network to a data lake. Eventually, after applying further analytics, the data will generate an alert of some form. At that point, the alert should trigger a workflow that raises a ticket in a service management system, and the ticket prompts a series of tasks, which involves any number of other systems that all have to work together smoothly and efficiently.

Are you dizzy yet? I am.

We haven’t even addressed the API layer yet, and it’s vitally important to consider the many API interactions that automation requires. APIs all have to talk to one another to ensure that IoT devices large and small, sensors, data flows, and systems are all operating at a high level. And in just-in-time manufacturing environments, all of the tasks and operations have to be timed perfectly.

So, when you consider the massive scale of industrial IoT, it’s easy to see why there’s a lot at stake. Worker livelihoods, corporate reputations, and literally billions of dollars are on the line every single day. And when you look at the intricate complexity of interactions necessary, it’s easy to see why the task is so daunting.

Automation Shortcomings

There have been three major shortcomings to date in the evolution of automation—untapped data, workflow obsolescence, and poor API management. Initially, we collected a lot of data that we simply didn’t do anything with. We’ve begun to address the issue with edge processing and advanced analytics, but we can still do better. Our second failure was setting up workflows and processes that quickly fell out of date. Because the cycles needed to constantly check and optimize these workflows and processes were so overwhelming, and because there was so much to automate, no one really went back to check things were working as they should be. Machine Learning and AI have promised to help out here, but this again needs a lot of work to set up and tend. Meanwhile, the API economy has exploded. Thanks to the Internet and cloud migration, the amount of integration involved is staggering. And we still haven’t quite come to grips with how to effectively manage the rapid proliferation of APIs and integrations.

This is a lot, I know.

Get the Overcoming API Complexity eBook

The industrial space is careening forwards in pursuit of automation flexibility and agility. But until we reconcile the scale, the complexity, and our dependence on an erratic Internet, we might as well be hamsters spinning our wheels.

It’s a new world out there, with a new enterprise reality, and it’s time we shift our focus from the IT infrastructure we control to the broader goal of ensuring seamless user experiences. That leaves us with this one unavoidable question.

What exactly are we going to do about it?

The short answer is, talk to ThousandEyes. The longer answer is to gain end-to-end visibility into all of these interrelated components that make automation possible, so we can test the performance levels of your applications and anticipate problems before they happen. And that is precisely what ThousandEyes makes possible.

Gone are the days of testing performance in silos. ThousandEyes offers IT teams comprehensive visibility into an increasingly complex environment that involves a growing number of clouds and delivers a greater degree of productivity. IT can now view the entire picture, with no gaps, to see and test APIs, networks, clouds, and applications.

ThousandEyes gives you end-to-end visibility, hop-by-hop throughout the connectivity, and user journeys through the applications and interactions with key APIs, whether you control them or not. ThousandEyes provides assurance that the networks are performing according to acceptable thresholds for loss, latency, and jitter and that you will be instantly alerted if they aren’t. ThousandEyes also gives you a correlated view of network performance and global Internet health so you can proactively address networking issues before they impact user experiences. It also lets you leverage a broader dataset through Internet Insights, which shows how local performance issues can be attributed to global events in core networks and services around the world. And ThousandEyes helps you ensure that your expensive and critical investments are working as intended. 

Peace of Mind for Your Connected World

However you view it, ThousandEyes enables you to see beyond just devices and data, and into the core components of IoT, including the Internet, cloud and multicloud environments, your network, APIs, applications, and into everything between. No other company has the breadth of visualization, the ease of correlation between events at an application or network level, or in the API, or the ability to set these events within a larger global context. Yes, ThousandEyes can help you do all of this. 

Pardon me while I take just a brief moment to exhale.

As enterprises continue to invest heavily in automation and IoT technology, their dependence on the Internet will only increase. This leaves these investments vulnerable to the Internet’s unpredictable outages and performance lags. But with complete visibility and deeper insights into network performance and the broader global Internet health, we can move past these obstacles.

With so much at stake, and so much riding on the Internet, ThousandEyes visibility and alerting can mean many things to your organization, and the most rewarding among them might just be peace of mind.

It’s time to experience for yourself just what ThousandEyes can do. Contact us to request a demo or a free trial of ThousandEyes Internet Insights, Network or Application Synthetics, or both.

Subscribe to the ThousandEyes Blog

Stay connected with blog updates and outage reports delivered while they're still fresh.

Upgrade your browser to view our website properly.

Please download the latest version of Chrome, Firefox or Microsoft Edge.

More detail