3 Tips for Building a Custom Dashboard in vROps

by bluemedora_editor on July 24, 2017

By: Scott Walsh

 

Custom dashboards transforms your vRealize Operations experience, but it can also be challenging if you have never built them before. After working with several customers to build dashboards, I’ve learned a few things that I’ll share in today’s post.

Tip 1: Remember what is important!

Don’t:  Forget that your dashboard is a problem solving tool.

It is easy to get caught up in the details of how widgets work, what data is showing, and how our visualizations lay out on the page. But sometimes it is important to step back and remember why the specific dashboard exists. Put yourself in the end users’ shoes, and think of it from their perspective.

 

Do:  Figure out what your users’ are afraid of.

What are users really concerned with. Here’s a summary of common customer concerns:

Costing – With a little (a lot actually) of help from VMware, we can help customers with their costing. Maybe we can tell the user when more capacity is needed now and a year from now. How could you find out which services are using the most resources? Is server {X} consuming more power than I want? Can I power {X} down or move {X} to low-powered hosts?

Uptime  – This is the biggest concern for obvious reasons, and it’s not quite as simple as “Is it on or off?”. Uptime is determined by the ability of {X} to service a request. Let’s say a host system is experiencing high storage latency. Is the host system overloading the storage or does the storage system have a saturated network interface and thus can’t reliably service a request. If {X} cannot service a request, {X} maybe not be ‘up’.  Many of our customers have Service Level Agreements (SLA) requiring them to maintain a certain amount of uptime. 99.9% uptime only gives you ~8 hours out of the year for maintenance and other follies. We can help a customer reach a SLA goal by providing insight and context to the stack.

Performance – Only a problem when something is not performing. I’ve often heard that diagnosing performance problems are like trying to find a needle in a haystack. The problem arises with the information given relating to the problem. “VM {X} is slow.” requires plenty of detective work before you can give a definitive answer to the problem. Dashboards can make this process much easier by providing a new context.

Scaling / Consolidation – How can an admin determine if they need to scale?  How would you know when you’ve reached the limit for a service? How could you identify under used resources in your environment? Can I fit more ram in a system without overloading the cpu? Can my storage keep up with a new rack? We can’t say “You need {X}.”, but we can give the customer more data to help them make a more informed decision. Interestingly, some of this question is born out of concerns for keeping costs down while maintaining performance. How much longer can I maintain performance? How much would {X} new hardware reduce my operating costs?

 

Tip 2:  Review your dashboards in the ‘real world’

Don’t:  Forget that your dashboards require context.

We can easily get caught up building dashboards for every scenario that comes to mind, but did you ever step back and view a dashboard with real data only to find out that it really isn’t that helpful? Whether you have peers or customers review your dashboards, don’t forget that without the proper context, your dashboards will not be impactful.

 

Do:  Show historical data when it is valuable.

One of the widgets that I have not leveraged enough, but am starting to love is the sparkline widget. Sparklines are a compact way to show historical data. When multiple sparkline widgets are stacked and on the same timeline, it gives you a nice picture of how multiple resources are performing over time. Do not underestimate a human’s ability to spot visual patterns. Showing a simple object list on the left, with a sparkline widget on the right can be impactful in and of itself.

 

Tip 3:  Learn what is acceptable in your environments

Don’t:  Make assumptions about your users’ environments and usage.

This tip is focused more on creating dashboards for users other than yourself. Obviously, we have situations as SysAdmins where we need to create our own monitoring dashboards. However, as the EngOps world continues to evolve, we find ourselves creating dashboards to share with various users and groups throughout our organization. Remember, they may have very different views of what acceptable performance is in their environment.

 

Do:  Review dashboards with users.

It never hurts to sit down with your users and take a look at the data in the wild. Maybe your development group has a different definition of ‘acceptable’ in development versus production. Maybe they are purposely performing load testing and require different thresholds to be set. They may even think that balancing the load across various servers is less important due to batch jobs that periodically run. The key is that you will not know what is important to your users unless you sit down and view the dashboards with them.

 

With these three tips, you can make your custom dashboards even more useful and ensure you are getting the most out of your vRealize Operations environment and the True Visibility Suite from Blue Medora.

 

Blue Medora Labs is the professional services arm of Blue Medora.  We provide custom integration with various components of the vRealize Operations Suite, as well as professional services utilizing more out of the box functionality.  We help our customers and partners provide a user experience which is more customized to their unique business challenges.

vRealize the most on your investment.  Get up and running fast with Blue Medora Professional Services.  For more information, visit http://go.bluemedora.com/proservices.

 

Get started

Take the quick 5-step assessment to see how BindPlane can meet your needs.

Subscribe
True Visibility
BindPlane for VMware vRealize Operations

True Visibility allows cloud management teams to use VMware vRealize’s powerful machine learning and capacity planning engine across their entire hybrid cloud environment.

SelectStar
BindPlane for the Data Layer

SelectStar makes it easier for DBAs and developers to monitor database performance during cloud transformation and DevOps adoption.

Thank you for contacting us. Your information was received. We'll be in touch shortly.