Making the Invisible Visible: Banishing a Network Ghost with Dual-View Observability

Halloween may only come once a year, but it felt like a much longer ordeal for the Hewlett Packard Enterprise (HPE) IT team, thanks to a network ghost that haunted us all year long.

Employees across 149 HPE offices worldwide would contact us about mysterious connectivity issues, including Wi-Fi that would cut out mid-meeting and applications that froze without warning. Whenever an IT technician started a session to troubleshoot, the issues would often vanish into thin air.

The phantom connectivity issues followed us everywhere. Since we had no way to verify our colleagues’ claims, troubleshooting was slow, expensive, and frustrating for everyone. Each incident would take up to eight hours to resolve, as IT technicians often had no other options but to sit on-site with the affected employee – sometimes all day – hoping the issue would reappear.

That kind of time commitment wasn’t sustainable for a team with thousands of employees to support worldwide.

A dark cloud over a large footprint

HPE is a multinational technology company supporting businesses around the globe with cutting-edge infrastructure and hybrid-cloud solutions. Within that larger ecosystem, HPE IT is responsible for supporting productivity within our internal workforce.

As a Dublin-based IT networks delivery manager for HPE, I oversee a team dedicated to keeping our office and campus networks working flawlessly. With a global footprint comprising approximately 7,000 devices, the scale of that responsibility is immense.

We knew we needed deeper visibility and more detailed insights if we were going to transform our network from a recurring source of frustration to something that consistently enabled our colleagues to maximize their productivity. And as a company specializing in top-tier infrastructure, we suspected that the solution already existed somewhere within our own portfolio.

An always-on solution shines light into network corners

We were right. It didn’t take us long to find the HPE Aruba Networking User Experience Insight (UXI) Sensors, small plug-and-play devices that act as 24/7 employee advocates, constantly testing a given network just as a user would. They connect seamlessly to Wi-Fi, proceed to open common business applications, and accurately measure end-user performance.

UXI sensors are like remote IT technicians that never sleep and can give us an ongoing, unbiased picture of network behavior. If there’s an issue or abnormal activity, the UXI sensors detect it immediately and alert the IT team — no frustrated user complaint required.

Once we completed the initial platform setup, adding a new UXI sensor to the mix was a simple five-minute process. All we had to do was plug it in, connect it, and within a few clicks, data began streaming to the dashboard. Suddenly, we could see what the user was experiencing, not just what they told us. We went from relying on anecdotal evidence to having concrete telemetry.

“Suddenly, we could see what the user was experiencing, not just what they told us. We went from relying on anecdotal evidence to having concrete telemetry.”

Today, we have around 800 UXI sensors installed across 120 of our offices, each one operating around the clock. They perform synthetic tests on some of our most heavily used applications and continually validate wired and wireless connectivity.

Since the UXI sensors run 24/7, we’re no longer reliant on chance timing. Instead, the UXI sensors “soak-test” each site, catching any blips or potential bottlenecks before they ever have a chance to impact a user.

Multiple solutions combat networking apparitions

The UXI sensors are only part of the equation. We also manage our global campus networks through HPE Aruba Networking Central, which gives us a top-down view of network performance based on infrastructure metrics and overall device health. Together with the bottom-up perspective from UXI, we have a comprehensive picture that includes both user experience and equipment behavior.

We use both tools side by side via the HPE GreenLake edge-to-cloud platform, which provides our engineering team with single-pane access to management and visibility. For example, HPE Aruba Networking Central might flag a switch that’s been running hot, while UXI sensors can confirm whether it’s affecting users. Conversely, the UXI sensors might pick up on slow application load times in instances where the associated hardware appears healthy. Instead of issues lingering for days until someone reports them, alerts occur immediately, and we can zero in on root causes in minutes, not hours.

In adopting these solutions, we’ve transitioned from reactive troubleshooting to proactive incident management. We integrated our UXI sensors with ServiceNow, which automatically generates tickets as anomalies are detected. We receive more support tickets now than we did before, but that’s a good thing. It means we’re finally capturing information from the furthest reaches of our network instead of performing guesswork based on isolated areas. And because every ticket includes details that allow our Global Network Operations Center to tackle issues much more effectively, our time-to-fix and time-to-own metrics have dramatically improved.

“Because every ticket includes details that allow our Global Network Operations Center to tackle issues much more effectively, our time-to-fix and time-to-own metrics have dramatically improved.”

Our next step is to extend this same level of seamless connectivity to individual employees’ devices. We’re piloting a UXI PC/Mac client that will run these same tests locally, providing us with yet another perspective on real-time user experiences. Once fully deployed, it will expand visibility to around 100,000 end users, helping us catch issues that UXI sensors alone can’t detect, such as snags related to laptop configuration changes or personal software updates.

The new HPE Aruba Networking Central dashboard also delivers new AI-driven insights, which are already changing how we work. The platform automatically recommends configuration improvements, highlights firmware vulnerabilities, and offers suggestions on which devices to upgrade. These details save us time, and we expect to streamline manual tasks even more with increased automation.

A haunt-free network experience

Our new HPE solutions gave the IT team observability with empathy, and by doing so, have banished our network ghosts. Connectivity simply works across 149 offices, thousands of devices, and tens of thousands of end users.

Along with our improved network visibility, the IT department has gotten a morale boost. Remote troubleshooting used to be one of the most frustrating parts of our job. Now, we can all clearly see what’s happening without needing to coordinate across time zones or interrupt anyone else’s workday, and we can focus on solving problems instead of chasing phantoms. It’s proof positive that the right visibility can turn even the scariest ghost story into a success story.

As a result, our team is much more innovative, efficient, and engaged. We’re proud to provide a truly frictionless experience for our colleagues so they can provide stellar products and experiences for our customers.