Cloud is scalable, it’s flexible, it’s a whole host of cool-sounding adjectives. But what does that all mean in practice? While it’s nice for the IT budget to be able to adjust infrastructure resources on the fly, cloud servers are also facilitating a subtle shift in the way your department should be managing infrastructure.
It’s time to quit relying on monitoring solutions and simply reacting to problems as they arise. Proactive cloud management involves digging in across every department in your organization to more closely align IT resources with business objectives.
Here’s how the new paradigm of proactive cloud management differs from simple monitoring, patching, updates, and other firefighting.
Cloud monitoring is still an incredibly important piece of any production environment. You need to keep a careful eye on network activity, firewalls, antivirus/antimalware tools, intrusion detection and prevention (IPS/IDS), storage capacity, and of course utilization rates of CPU and memory for each virtual machine.
Built-in monitoring with vSphere and additional tools like CA Unified Infrastructure Management or other third party platforms like Dynatrace will let you generate reports and alerts to achieve your monitoring goals. Many cloud providers also offer monitoring as a service, with reports sent directly to you and with any potential issues noted.
Cloud monitoring should be at least a regularly scheduled scan, if not a constant scan, that provides a summary of any changes in the baseline environment.
This longstanding mode of monitoring must remain – you don’t want an application to fail due to performance issues or malware.
Monitoring does play a key factor in proactive management practices. Continue keeping a close eye on your cloud infrastructure and resolve issues as required. But instead of being constantly reactive and then putting out fires, start laying the groundwork for a collaborative, proactive IT management style that works across your entire organization.
The main shift here is going from resolving problems as they are presented, working with existing infrastructure, to predicting future demand and obstacles and adjusting your virtual datacenter in anticipation.
There are two main benefits to this style of management: you discover IT demand that you may not otherwise have known about, improving the image of IT within your organization as a value driver rather than a cost center, and you avoid potential VM failure from resource contention or security holes.
Costs can pile up if you aren't using AWS for the right scenarios. Download this white paper to learn more.
Patching and upgrade management is likely already being practiced by your department, but if not, now is the time to lay out a regular schedule for patching and upgrades for software and firmware, even for machines that should be up to date. Stay abreast of potential patch releases so you can work them into change management as soon as possible.
The scalability of cloud plays a vital factor in proactive management. When do you need to scale up or down? Predictive deployment requires you to track utilization rates over time to see when you might need to add or remove capacity. Are your storage IOPS keeping up with your VMs? Will they after you have a few hundred additional gigabytes of data traversing the network? Project your current workloads out for a year in advance. Then multiply it by 1.5. That’s where you need to plan to be. Take steps toward it as budget and staff levels allow. If you overprovision, you can always scale back.
In addition, stay in regular contact with department heads to learn when increased IT demand might be coming down the pipe. Are any large initiatives planned that will stress IT infrastructure? Don’t wait to react to an order for more capacity. Go out and find it.
As long as you’re scheduling regular meetings throughout your organization, ask how IT is meeting expectations for software and services. What new applications or abilities could be provided? Are there any web services in use that you don’t know about? Being proactive is the best way to fight shadow IT.
Proactive management goes all the way down to small efficiencies, each of which can add up to fight resource contention. Can you disable unnecessary services within your server OS, for example?
Have you already been practicing proactive management? Tweet your favorite ways to get ahead of IT infrastructure problems and we’ll feature them in a future post.