Peter Drucker said that when you walk into a poorly managed factory, it seems like a lot is happening. But when you walk into a well-managed factory, it seems like nothing is happening, because all the problems have been anticipated and dealt with. Unfortunately you can’t just “walk into” your virtual environment and ensure that everything is unfolding as it should.
Just like car manufacturers for many years had an inviolable rule “Never stop the line,” the modern IT equivalent is downtime.
Avoiding Downtime Requires Anticipation, Planning, and Oversight
Downtime is very bad news for any business. Even if you don’t have any customer-facing assets, like a retail website, almost everything that your employees are doing most likely relies on IT assets like networks, email servers, and other applications. In some cases, they might not even be able to use the phone! IDC has released a report that shows that for a Fortune 1000 company, the average hourly cost of an infrastructure failure is $100,000 and the average hourly cost of a critical application failure ranges from $500,000 to $1,000,000.
In spite of the enormous costs related to IT problems, too frequently, insufficient resources are directed toward the planning and monitoring of various systems and infrastructure. Sometimes fault reports are even turned off because they cause too much signal noise! And sometimes the monitoring is simply outsourced, but the relationship has too little transparency and too much distance to be effective.
Design and Maintenance are Equally Important
Effective design of network monitoring is not trivial. That is, it is relatively simple to diagram, but as IT assets become increasingly diverse, it is not an easy task to ensure that the Cisco box talks to the IBM box and the virtual machines are all identified and and the VMware has all the kinks worked out, and that there are sensors everywhere there needs to be sensors and that they are all working to spec. Network administrators earn their keep.
Once everything is plugged in and able to be monitored in the hypervisor, the next step is to devise ways of making sense of all the information that is suddenly flooding towards you. Is it helpful to see the SQL server’s CPU usage on the same screen as the web server? Probably not; therefore there needs to be an intuitive structure to the various screens as you review the landscape. If the ping to the firewall fails, do you really need to get another alert every minute until it succeeds? Probably not; therefore, there needs to be a set of notification rules in place to avoid a distracting cascade of fault reports.
Once this is set up, the work doesn’t stop. As IT deployments are constantly shifting, each new change has to be plugged into the system with the same attention to detail as the first in order for network monitoring to remain robust over time.
You’re Not Just Paying For Monitoring; You’re Paying For Response Time
A beautiful setup and an intuitive hypervisor is a great thing to have and arguably the very first step to avoiding all kinds of IT problems.
But once again, you fall into that most common of technology issues: Whose job is this going to be?
Monitoring is all very well. But it’s three A.M. and your sysadmin is at home, fast asleep. Most companies have trouble finding enough IT staff for the ordinary 9-5, let alone three shifts’ worth. You need a monitoring service to ensure that when there is an issue, it is dealt with immediately, and effectively, by experts.
On the other hand, finding reliable and cost-effective monitoring services is not easy. However, when you find a company with both the expertise and infrastructure to monitor your IT assets, you will greatly shorten the process of getting everything ‘on the same page.’ After all, they do this sort of thing on a daily basis, and over time have learned all the tricks of getting heterogeneous equipment talking to each other. Now that you can’t rely on everything being IBM, this skill cannot be under-emphasized. Having saved all that money on white box hardware, you can easily lose it on the back end if your staff spends all their time troubleshooting.
Let us worry about your network monitoring. From design to setup and all the work that come after, our network management services will set your mind at ease.
Colocation: The First Step to Private Cloud?