Infrastructure monitoring vs. NPMD (network performance monitoring and diagnostic)

What is IT infrastructure monitoring and how does it compare to network monitoring? In this post I’ll be focusing on IT Infrastructure monitoring and how is it different than other types of popular monitoring categories.

That moment when you are compared against someone else…

I work for LogicMonitor as the VP of Product. LogicMonitor provides a SaaS-based hybrid IT infrastructure monitoring. Our platform uses automation to detect and monitor all resources on-premises, in remote data centers and in multiple clouds.

Last month, LogicMonitor was placed on the Gartner Magic Quadrant for the first time in the category of Network Performance Monitoring and Diagnostics (NPMD — See the original news post). Given the reach and influence that Gartner carries, this is exciting news for us — especially since network monitoring is not our only focus. We were placed within the quadrant of “niche player”. It is a respectable placement — the highest within that quadrant on one of the two axes. But it makes me wonder whether there’s a common understanding of infrastructure monitoring and the advantages it offers beyond network performance monitoring.

Categories are limiting

The following companies come up in reference to the term ‘infrastructure monitoring’ when googled: LogicMonitor, Dynatrace, New Relic, Oracle (Cloud), Sensu, AppDynamics (calling it infrastructure visibility), site24x7, Poweradmin, OpsView, Splunk and others. Infrastructure monitoring is a common term. Gartner defines it as a tool “which tracks all the components and performance of your IT infrastructure. The smart monitoring systems capture the availability of various infrastructure components residing in a data center or hosted in the cloud as infrastructure as a service (IaaS). These tools monitor and collate the availability and resource utilization metrics of a server, networks, database instances, hypervisors, and storage.” Gartner is also using that categorization in a site called Gartner PeerInsights.

Why invest in IT infrastructure performance monitoring?

Based on ten years of experience working with more than 1,500 customers, here’s what we’ve learned customers need in an infrastructure monitoring platform:

• Smart monitoring of the health, availability, utilization, and performance of everything in the on-premises and cloud data centers — beyond any one IT silo. These components include physical or virtual servers, all networking elements, network traffic, database instances, hypervisors, and storage systems performance.
• Coverage is key: when choosing an IT Infrastructure solution, users want a solution that covers all (if possible) or most of the devices and systems that they have. Being able to support things beyond typical IT equipment like basic telephony and other IT needs is a plus
• Hybrid: users want to manage everything they have using one unified view (pane of glass). Having a solution that only supports cloud (not to mention a single type of cloud) or only on-premise no longer makes sense.
• Flexibility and extensibility: no single solution would cover everything out of the box. Solutions that are easily extensible and allows users (or 3rd parties) to add coverage to things that are not covered out of the box is important. The extensibility is not limited to any specific type of interface or technology. It can encompass any system, resource or device that emits data, like an industrial freezer, sensor, camera, etc.
• Availability: in order to trust the monitoring system, users want to ensure uptime and availability. This can be achieved by built-in redundancies, like operational independence from just one public cloud provider (in the spirit of “Who will guard the guards themselves?”).
• Fast time to value: one of the important things users want to minimize is long deployments over a large number of device types. Ease of deployment correlates to faster time to value.
• SaaS: many customers prefer a SaaS solution because it enables easy deployment, provides inherent redundancy, and reduces total cost of ownership (less maintenance and support than on-prem solutions)
• Smart alerting: users monitor their environments to know when something is not working as expected. When that happens, they need their monitoring solution to provide the most concise information about what happened and what needs to be done about it. However, it is important for a solution not to flood the user with redundant alerts. Instead, it should provide the least number of alerts that the user needs to be notified of a situation and its context.
• Topology and relationships: users need to understand the relationships between the physical and logical elements in their environment to enable faster troubleshooting and visualization. These relationships can be multi-layered and of multiple types and domains.

Beyond the things that customers require today, there is a very important aspect that pushes IT infrastructure monitoring into the future, and that is the fact that it keeps evolving with the on-going changes applied to IT. With cloud computing, containers, virtualization and the move to software-defined infrastructure (networks as well as storage), IT Infrastructure monitoring keeps evolving to address these new domains. The constant changes driven by disruptive technologies are pushing customers towards a new generation of tools within the IT service management domain.

And how does that contrast with NPMD?

The Gartner definition for NPMD is “tools [that] allow for network engineers to understand the performance of applications and infrastructure components via network instrumentation. Additionally, these tools provide insight into the quality of the end user’s experience. The goal of NPMD products is not only to monitor the network components to facilitate outage and degradation resolution but also to identify performance optimization opportunities. This is conducted via diagnostics, analytics and debugging capabilities to complement additional monitoring of today’s complex IT environments.”

Products in that category usually have:

• The ability to monitor diagnose and generate alerts on network endpoints(everything that has an IP), components and links, end-user experience, business service delivery and infrastructure component interactions with the network.
• Provide analysis for real-time and historical performance behavior
• Utilize information sources like NetFlow, SNMP and packet analysis
• Scalability to support 10G Ethernet and be able to ingest and analyze large amounts of data per second.

It’s clear from the above definition that NPMD is highly focused on the network side of things. But his definition also exposes a limitation: a network monitoring solution is less relevant in a cloud environment, where most of the network is abstracted away. So network monitoring solutions gravitate toward on-premise environments. In addition, network-related protocols that are used (SNMP, NetFlow, and packet analysis) may be limiting in a wider scope where additional protocols may be required to bring in the necessary data.

In other words, some IT infrastructure monitoring platforms can be considered network monitoring solutions (such as LogicMonitor), but most network performance monitoring solutions are not broad enough to cover IT infrastructure effectively.

Conclusion

IT Infrastructure monitoring is a significant superset of network performance (from the point of view of the customer) in terms of areas that can be covered and protocols and APIs that are typically used. My observation is that infrastructure monitoring tends to follow more tightly on new technologies and be more disruption-resilient. I believe that for NPMD solutions there is an expectation that the networking related capabilities be very comprehensive. In that sense, it makes sense that in NPMD playing field, LogicMonitor is placed as a niche player. LogicMonitor is covering many additional domains and is not focusing only on the networking side — because we believe that the value provided by a comprehensive view of infrastructure is greater than a deeper, purely network view.

There is a wider scope that is IT Infrastructure monitoring. We are playing in that category very well. Our customers are attesting to that and hopefully one day Gartner MQ would have such a category. We are very happy to be on the MQ of NPMD and it’s likely that we will be in a better place in the next MQ review, even if it is just a small subset of the value that we provide for our customers.