There is a quote that almost everyone working in a management role will recognise: “if you can’t measure it, you can’t manage it”. Naturally enough, IT is no exception. After all, if you haven’t got good visibility of how the data centre is running, how can you ensure it keeps functioning effectively, and efficiently? This makes the title of a recent research note by Freeform Dynamics, The monitoring capability gap, something of a concern for data centre managers.
Starting with the basics, the report makes it very clear that the monitoring tools used by respondents – and maybe you – to keep systems and services operational have plenty of scope to be improved. (Figure 1).
The first thing that leaps out from the results is that half of respondents say the monitoring tools they use do not provide all the functionality they need, twice as many as those who say that they do. Even if everyone always wants more (and we do), the other responses reveal other issues that need attention. For example, significant numbers of those taking part report that their monitoring capabilities do not allow the speedy identification of the root cause of a problem.
Even though a majority of those taking part in the survey agree or strongly agree that their tools provide good end to end visibility, almost a quarter disagree or strongly disagree. Indeed, almost one in three say their tools do not allow them to proactively manage systems, a matter of concern when data centre complexity is increasing, and as line of business users become ever more intolerant of IT problems or slow-downs, since so many business functions can no longer operate without IT. Add Cloud and other external resources into the mix and it is easy to see why many IT teams sometimes have to battle with monitoring and diagnostics.
Data centre and infrastructure monitoring have traditionally been at the heart of IT operations, and the results show this is very much still the case. (Figure 2).
Business users have obviously benefited greatly as computer systems have become ever more reliable, and as data centre managers have built systems with availability a primary concern. But while resilience was once the preserve of just a few systems, the portfolio of applications and services regarded as critical has now expanded dramatically. This explains why, despite the advances in core IT technologies, monitoring the health of systems and components still consumes considerable time and resources. When combined with the fact that critical business operations rely on IT systems not just to be available but also on them delivering excellent performance, it is clear that the need for accurate, sophisticated monitoring is not going away any time soon (Figure 3).
The results highlight the widespread distribution of the challenges faced by data centre professionals today. It is very clear that in every area – even in the fundamentals of problem identification and system visibility – there are usually more survey respondents finding things difficult or very challenging than there are who report no challenges.
What is less obvious to anyone outside of IT is just how important monitoring tools are in these matters. This may well reflect that unless you operate monitoring tools, day in, day out, it is very easy to overlook just how significant they are. It is even true that some IT professionals can take things for granted, especially if they have built complex scripts and daemons to help with some routine functions.
The IT landscape is changing dramatically and rapidly, innovative technologies are arriving in the data centre every year with cloud systems becoming important to an expanding range of business services. And then there are new architectures for developing applications, such as containers and serverless, that are beginning to hit the mainstream. Taken together these make the results at the bottom of the chart perhaps the most alarming. Despite the changes mentioned above, over half of those taking part report it is at least quite difficult, or even very challenging, to get agreement to upgrade monitoring capabilities to cater for the new IT landscape. Even more of them report that it is hard to get budgets to fix things that are sustainable in the long term.
But can anything trigger investment for monitoring capabilities (Figure 4)?
Perhaps unsurprisingly the events that are likely to have the biggest influence on triggering investment in monitoring tools are anything but desirable. Even in a world where customer trust has never been so important, and the attention paid to data security is gathering momentum, the event most likely to get the business ready to spend money is a significant security breach, with significant systems outage only slightly behind.
It is interesting to note that far fewer respondents stated that a big security scare in the press would have the same impact. With GDPR a reality from May 25th, 2018 and the potential for penalties of up to €20 million or 4% of worldwide revenues, I suspect that once a large fine has been handed out the first time, the number of organisations investing in monitoring technologies may well increase dramatically.
The Bottom Line
Every manager and IT professional working in a data centre understands just how important IT is to keep the business running. This makes monitoring technologies more relevant and essential than they have ever been. The complexity of IT systems, the critical functions they support and the stress under which they must operate are all increasing the need to ensure systems run effectively.
At the same time, financial demands dictate they be run effectively – underutilised IT resources are becoming a luxury of the past. Monitoring is essential, but has been underappreciated. Yet if you can get it right, things are more likely to work well; conversely, if you neglect it, the negative impact for the business will be noticed.