VMware vSphere best practices: Why you need them


You probably already know that VMware vSphere is used in most datacenters around the world to run critical applications. But what you might not realize is that vSphere is often implemented with default configurations and standard features. And, once deployed, many vSphere implementations are not regularly evaluated to determine potential improvements in terms of performance, security and advanced functionality. It's almost too easy to click the checkbox that enables vSphere high availability—which should be a signal that it’s not enough. You should also ask important questions, like:

• Were the requirements met before enabling it, was the failover capability tested (to ensure that it works)?

• Were advanced availability options considered? (For example, guest OS monitoring?)

• Were best practices for network performance and availability followed and has the network been analyzed lately for optimal configuration?

• What about security and data protection?

• Are there any performance troubleshooting or capacity planning tools in place?

These are just some of the things that need to be asked about any vSphere infrastructure, especially one that's expected to run your critical applications with high availability, optimal performance, and strong security. Now, let's take a look at what else you might need to know about vSphere best practices and how you can begin implementing them today. 

vSphere best practices: Why bother?

I'm sure there are many vSphere admins with immaculate vSphere infrastructure. These are the folks who've spent significant time understanding requirements, planning  their design, reviewing best practices, testing performance tweaks, securing their infrastructure, and right-sizing their VMs to the needs of their applications. But in most cases it’s likely that the vSphere admin also wears 12 other hats and barely has time to keep things running. 

The thought of taking a week to review vSphere best practices for security and performance, across a medium or large infrastructure, is the last thing on their mind. In many vSphere infrastructures, a consultant (with good intentions) may have been brought in to install and configure vSphere--and then may never have been contracted to return. Over time, that vSphere infrastructure has changed and, without a full-time administrator, its performance and security is probably lacking.

Let's frame it another way: Think of vSphere best practices like a car seatbelt. You’ll probably survive a few days, weeks, or even years without it—but sooner or later the unexpected happens, leaving you wishing you’d used it. vSphere best practices are similar, as you may be fine for a period of time (yes, even years) before something goes wrong but, when it does, you'll be kicking yourself for not taking precautions earlier on.

Think about this for a moment: What would be the impact of all your VMs going down? Proper analysis for best practices can help prevent that costly nightmare.

Data Center best practices: Which types should I use?

In general, the types of best practices to implement in vSphere are those that you should implement in any traditional datacenter infrastructure. For starters, you should always look at best practices for performance around compute, storage, and networking.

Other common best practices that you need to be aware of in virtual or physical environments are related to:

  • Management & monitoring
  • Availability
  • Security
  • Performance
  • Data protection

As with traditional datacenter sizing, virtual infrastructure sizing works the same, to a point. When it comes to compute, storage and networking, it's all about ensuring that there’s enough throughput and capacity for workloads (both for today and for the future). However, with a virtual infrastructure, you also must ensure that your virtual resources (provided by the hypervisor layer and assigned to each virtual machine) are allocated correctly for the needs of the critical applications. Many people call this rightsizing.

Unlike in an all-physical environment where you typically have one application per server/host, with a virtual environment you’re trying to maximize the datacenter investment and run as many applications as possible on the same hardware. That means that you try to push up the utilization without over utilizing your capacity and causing application slowdowns.

This is a fine line to walk, and VMware offers some tools (like clustering, resource pools, memory optimization, and distributed resource scheduler) to help you achieve that. Rightsizing virtual machines for dynamic workloads and managing workloads across a cluster for optimum performance are things that must be constantly monitored, and frequent adjustments must be made in dynamic environments. While you can accomplish some of that with vCenter, many companies use tools like vRealize Operations to get more functionality.

With other areas like management, monitoring, availability, security, performance and data protection, there’s a lot to know (more than we can cover here). For this, you may need to tune into my new vSphere Best Practices course

Best practices: How will they really help me?

Whether you’re implementing best practices in home security, holiday shopping or vSphere infrastructure, your goal is to gain the best results possible. Best practices come from people who’ve been there before. People who have learned through trial and error. In other words, they made mistakes so that you don't have to.

Let’s go back for a minute to the seatbelt comparison. Seatbelts were invented and mandated as the best practice (and later as the law) by people who had been hurt (or had known someone who was hurt) during a car crash. Apply that example to vSphere and it’s true that vSphere best practices were created by people who had their vSphere infrastructure (and critical applications) die. Wanting to avoid this in the future, they came up with some best practices to prevent the worst-case scenario. Thankfully, they shared those best practices with the world. In the case of technology, many best practices come from technology vendors and consultants—but, in some cases, they come from regular vSphere Admins who just want to help others.

To sum up vSphere best practice benefits, they ensure that your vSphere infrastructure:

  • Runs as efficiently and optimally as possible
  • Is as secure as possible
  • Is as high performing as possible
  • Is as highly available as possible

How do I learn vSphere best practices?

You could try to learn vSphere best practices via trial and error on your production infrastructure (if you want to risk losing your job). Or you could read hundreds of pages of boring and complex best practices documentation from vendors. Thankfully, I’ve got a third option for you that will make this all much easier and far less time consuming: Check out the trailer below to know more about my new vSphere Best Practices course, and learn what you need to know to keep your vSphere infrastructure highly available, performing well and highly secured. 


David Davis

David Davis has authored over 50 courses for Pluralsight around enterprise data center technologies such as cloud computing, virtualization, and (especially) VMware vSphere. He is a partner at ActualTechMedia.com where he creates compelling enterprise technology content, moderates online events, and helps to connect some of the best-known technology companies in the industry with the end user community. With over 20 years in enterprise technology, he has served as an IT Manager, administrator, and instructor. David is an 11x VMWare vExpert, VCP, VCAP, & CCIE# 9369.