Well, Isn’t That Convenient? – The 4 Cs of Quality IT Monitoring Tools

Some things in life are easy and some things are hard. Sleeping-in on weekends (provided you don’t have small children), shifting into 6th gear and ‘flooring it’ on a deserted stretch of Eastbound I-70, and taking a swig of Breckenridge Brewery’s Vanilla Porter go in the “Easy” column.

What kinds of things are on the “Hard” side of the ledger? Let’s keep it succinct and say it’s the opposite of everything I just mentioned: Exhaustion (I have small children), bumper-to-bumper traffic, and drinking Keystone Light beer. Something else I’ve noticed is difficult over the last 20 years is systems and network management. Day-to-day IT administrators are tasked with keeping track of myriad technologies, numerous management tools for said technologies, and the demands of managers who just want usable data for decision making purposes. It is the job of NMS software vendors to make the difficult tasks of network and systems management if not easy, then at least more convenient. Continuing with our “4C’s” blog series  we get to the next attribute of quality IT monitoring tools: Convenience.

What does it mean for an IT monitoring tool to be “Convenient”? Simply put, the tool needs to be easy to implement and use. In my travels with customers and prospects alike, it’s not uncommon to find only a small handful of ‘Jack-of-All-Trades’ IT personnel responsible for vast IT environments. To that end, you want to find tools (monitoring and otherwise) that are convenient to setup and maintain. Do these tools contain at least some measure of automation capability, like network auto-discovery, so your IT staff isn’t spending all their time adding and subtracting devices?

Aside from automation, how well does this tool scale to a large environment? Is your team configuring devices one at a time or can the entire configuration be done with a handful of points, clicks, and selects in your browser window?

Another incredibly important aspect of a tool’s “convenience” factor is its expansiveness. Are you limited to interacting with the tool through only a web browser interface? Does it provide full API access so you can communicate with it programmatically? Is the tool well documented for the DIY-inclined? If you can answer in the affirmative for most of these questions, then you’re on the right track.

Of course, everybody wants things to be easy and convenient. However, it’s crucial to understand why it’s important to make tool use and configuration convenient. I suppose once in a blue moon you’d find an IT Director or CIO out there who’d sign a P.O. worth potentially tens of thousands of dollars for an NMS system with a justification of “… because it makes my job easy” from their engineers. However, that’s not a particularly likely scenario.

To analyze the “why” in this situation let’s break things down from both internal and external perspectives with respect to a typical organization. From an internal point of view, an easy to use NMS system is a tool that increases your staff’s efficiency. What metrics would you use to judge IT efficiency? The most common is ‘Mean Time To Repair’ (MTTR). When a there’s a service outage how long did it take from the time the incident started until the time the service was restored? The intermediate steps of troubleshooting and root-cause analysis should decrease (which is equivalent to an increase in IT efficiency) with applications that are easy and convenient to work with.

The same metric of MTTR can be considered from an external or customer-centric perspective as well. However, a stronger measure (linked to MTTR, by the way) is your company’s bottom line. If your service or product is hampered by outages, poor performance, or other technological maladies, then word will get around quickly. Your customer base will melt faster than a pat of butter on a hot frying pan. Convenience and ease-of-use are indispensable to maintaining and increasing IT service levels.

Okay, so we’ve got it  covered what “convenient” really means. We’ve examined at least one metric for quantifying why it’s important. Now we want to ask “How do we get there?” What are some tangible attributes of quality NMS tools?

Scale, Scale, Scale

Look for an NMS tool that can handle large quantities and a wide breadth of IT infrastructure equipment. All flavors of networking gear and servers are a no brainer, but also look for something that can automatically discover and monitor SD-WAN technologies and virtualization platforms (among other things). Can one core NMS system scale up to handle thousands upon thousands of devices? Similarly, does it allow you to quickly configure said devices? Jack-of-All-Trades types don’t have time to point and click through a user interface for days upon days to get their devices monitored.

We Don’t Need No Stinkin’ Agents

Go for an agentless vs. agent-based monitoring solution. Yes, I acknowledge that going this route does mean sacrificing richness of data from your monitored devices. So what? The name of the game here is convenience, ease-of-use, and ease-of-configuration. It’s time and resource consuming to mess with installing agents on any system you want to get performance stats from. You might be thinking “… but I have {enter name of orchestration tool here} so pushing out software and configurations is easy.” Again, so what? WMI/WinRM is a pretty complete diagnostics framework for Microsoft Servers and on the Linux/Unix side of your server infrastructure Net-SNMP is completely extensible. You’ll get probably 80% of what a locally installed agent program can provide. Additionally, if you do have to hunt down a problem you’re most likely logging into troubleshoot the device anyway to address the issue. Save yourself the hassle and computing overhead by shunning any notion of an agent-based toolset.

Skynet Here We Come

The new hotness in performance monitoring and alerting are tools that take advantage of machine learning algorithms in your monitoring strategy. I’ve written about this topic previously so I won’t rehash it in this article. Suffice it to say, if your chosen tool does take advantage of these types of algorithms and they come configured out of the box with sensible defaults, then you’ve got an immediate ally in your monitoring strategy. Historical data can immediately begin to be analyzed for patterns and then alerted upon.

Stretching the Boundaries

Last, but certainly not least, is the expansiveness of the tool. Not unlike the utility you’d find in pants with an elastic waste band after Thanksgiving dinner seek out a tool that expands how you can interact with it. A great example here is API access. If you’ve got a complete API into your network management system, then you also have the ability to tie it into your other mission critical applications.

Also, look for a tool that has a robust mobility component. The nature of NMS systems is that they do contain sensitive data, and opening it up for access remotely can be complicated. If Dorothy and the Scarecrow were network engineers and charged with this task I can envision them singing “DMZs, Firewalls, and ACLs oh my!” If it’s too cumbersome to setup remote access to your NMS, then it’s a project that won’t be completed. Choose a vendor that offers a better path forward with regard to mobile device connectivity.

Enterprise Infrastructure Monitoring is hard, and IT organizations face considerable challenges. False alerts, fragmented views of performance data, and rapidly evolving technologies are only the tip of the iceberg. The last thing any administrator needs is a management tool that adds to that list of hurdles. Given the information in this article, hopefully I’ve helped give some clarity on what attributes should be focused on to make that task easier. Now, if you’ll pardon me, I have to polish off this Vanilla Porter and go take a nap (since my kids woke me way too early this AM).

Ready to get started? Get in touch or schedule a demo.