Loyola University of Maryland lowers administrative overhead with unified observability

Challenge

Loyola University of Maryland, a Jesuit Catholic university located in Baltimore, Maryland, is home to 5,500 students across the main campus and three remote campuses. The Technology Services Department is the centralized provider of technology to the entire Loyola community, supporting applications, storage, security, connectivity, and all devices.

The challenges of so many devices, combined with the importance of uptime and visibility, led Loyola to search for an observability platform after a radical redeployment of the university’s network. What started as a simple hub-and-spoke topology, where all services symmetrically overlaid the network elements themselves, was migrated to a flexible collection of virtual networks operating independently of each other to serve different constituents and functions.

Because of this major network change, as well as an impending rollout of VoIP and the need to support new network-first facility safety devices, Loyola soon realized the mentality and capabilities of the native monitoring tools they had been using were not going to be enough.

“We just don’t have the time to be babysitters. We can’t afford to spend our time keeping the lights on. That’s a recipe for disaster,” called out Mike Dieter, Senior Systems Engineer in the Technology Services Department.

Loyola needed to minimize administrative overhead and day-to-day demands of an observability platform, as well as persistence in business continuity and disaster recovery type situations. Visibility was key to Loyola. If some or all of their services were disrupted, they didn’t want to waste precious time trying to bring those back up without any visibility into their environment.

Loyola was also looking to the future when it came to its partnership with an observability platform. They were going to need to easily access not only their data but also their account team to make any changes to their evolving and growing landscape. Overall, Loyola needed a partner that would grow with them.

Solution

Loyola knew they had found the solution they needed with LogicMonitor’s SaaS-based platform, out-of-the-box interoperability, flexible alerting, and notification delivery.

“Nothing is more frustrating than wasting time struggling to get somebody’s product to talk to somebody else’s product,” said Dieter.

LogicMonitor’s robust integrations have allowed simple and fast discovery, monitoring, and visibility of Loyola’s infrastructure, including Juniper, Palo Alto, and Aruba devices. Within minutes, Dieter and his team can see reports, metrics, and data points within their dashboards, with ease.

Loyola has also leveraged LogicMonitor’s integration with third-party notification delivery mechanisms, giving peace of mind that when issues arise, the correct person will be alerted and provided the information they need to quickly resolve the problem.

Loyola has partnered with LogicMonitor’s Account and Professional Services teams throughout their relationship. As their use case changes and evolves over time, they are able to add new features and functions quickly that align with their vision.

Benefits

Loyola was able to lower its overall administrative overhead with SaaS-based collectors, allowing the Technology Services Department to focus on data-driven decisions to advance their goals, rather than just trying to keep the lights on.

“It’s a huge improvement over past experiences with other products. Now it frees up our time to spend on things that add value back to our operations, instead of trying to figure out why this isn’t working. All of that is gone and we’ve really reaped the benefits,” Dieter said.

By partnering with LogicMonitor, Loyola has been able to provide and support connectivity across their community, with key insights and visibility into their growing virtualized environment.

In 2018, Loyola worked with LM’s Professional Services to implement a custom datasource to track switch ports in wiring closets. Years of renovations, office relocations, and the trend towards WIFI resulted in a significant number of abandoned wired connections across almost all buildings.

After observing the datasource output for an entire academic year, the team was able to identify and quantify closets where port capacity was no longer in balance with actual port usage, resulting in the removal of nearly forty switches from production.

“his was really cool for us and had a positive impact on our electrical consumption and carbon footprint. Electricity didn’t completely go away obviously, but we can certainly justify $5000 in cumulative electrical savings by now. Switches run around the clock for maybe 5-7 years, which adds up quickly,” Dieter explained about the high-value return provided by this engagement.

In 2020, the University’s Azure deployment reached a critical mass and Loyola adopted LM Cloud, gaining side-by-side observability for both on-premises and cloud infrastructure. In early 2022, the platform they were previously using for log collection no longer matched their use-case demands and was susceptible to a nascent security concern. The team was able to solve two problems at once with the adoption of LM Logs, which was not vulnerable to that security concern and provided the needed log visibility.

Tool consolidation produced consistency for Loyola, who no longer had to deal with having multiple alert mechanisms. They are now able to effectively administer who gets notified for what alerts and when saving many hours of administrative time.

“With consolidation, the big benefit is for multiple teams to have consistent visibility across a range of technical infrastructure. When we have to talk to each other, we’re now comparing apples to apples because LogicMonitor is giving us consistent metrics for reference points,” said Dieter.

Reducing tool sprawl with the adoption of LM Logs also meant savings costs for Dieter and his team, who saved as much as $2000-$3000 a year by consolidating. The total of other alternatives would have been as much as $10,000 a year spent to keep the platform current and in production.

“For the University, IT infrastructure is evolving far faster in 3-5 years now than it did in the previous 5 years. What we really like about LogicMonitor is that it gives us the flexibility to adapt to those changes as it occurs,” said Dieter.

Loyola has leveraged the flexibility of LogicMonitor as the industry shift from on-premise to cloud happens quicker and quicker. The demands Dieter and his team are facing are changing faster than ever before, so the ability to add more to their arsenal when needed, like with the adoption of LM Logs, is critical.

Looking forward, the Technology Services Department is actively looking to structure and keep pace with the growing capabilities of LM Logs, including tripping alerts and notifications on log events and using dashboards to visualize log event patterns.

Loyola also has a huge project in place to replace switches in all of their locations that are end-of-life in the next two years. “Using LM Envision, LogicMonitor’s platform,” Dieter said, “is a very tangible return on our investments.”

Logs provide new improvements

The Situation

In September 2022, during the first week of school, one of Loyola’s WiFi controller nodes suffered an event that caused it to lock up and stopped responding to SNMP DataSources and any attempts at remote access. The team didn’t notice this issue immediately as High-Availability implementations can often mask the exposure of service delivery issues by standard SNMP monitoring. 

The Solution

By utilizing LM Logs, the team was able to realize the extent of the situation. While the device wasn’t responding to SNMP DataSources, it was creating a significant volume of syslog events causing the team to immediately initiate a call with Aruba Technical Support. Senior Systems Engineer, Mike Dieter, said “before they were even able to get a case open, I had downloaded a short time-slice of logs from the start of the situation and presented it to them.” It took Dieter 90 seconds to enhance an existing saved log search, customize the time frame, download the resulting file and share that with his colleague for delivery to Aruba’s team to quickly troubleshoot and resolve the issue. 

A week later, the other WiFi controller node failed, but this time the team was prepped for quick resolution. Based on the previous occurrence, Dieter had created a Logs pipeline and a corresponding alert that tripped and notified as designed. Dieter commented, “LM Logs worked perfectly and how I could imagine the product managers envisioned.” 

The Results

Comparing the start-to-finish time frame of the two incidents, LM Logs improved Loyola’s response time by over 2 hours. Dieter explained this was just one example of many other times that he and his team relied on LM Logs with immediate alerting and notification engineered to troubleshoot and rapidly resolve issues before they affect the university and the students studying there.