SHAMELESS PLUG - So What's the Big Deal with vCharter Pro/Foglight for VMware?

Written by Scott Herold

Saturday, 29 March 2008 21:44

Those that attended VMworld Europe in Cannes this last year and stopped by the Vizioncore booth were treated to more than simply being able to watch George Pradel at his peak. They got to see, first hand, the new Vizioncore vCharter Pro. I know what you are thinking..."Pro" means a few features were added and a few bugs were taken care of that users have been complaining about for months. Well, if that is what you are thinking, you actually couldn't be more wrong.

If there was a way to envision taking a piece of software, covering it in gasoline, lighting it on fire, and throwing it into a cluster of highly explosive material, you can start to grasp at what happened between the vCharter and the vCharter Pro releases from Vizioncore. The original version of vCharter was a unique product when it first launched. It took ESX Host data and displayed it in a way that was easy to understand and allowed for an administrator to very quickly determine root cause analysis of things that didn't look right. It was a very easy to use troubelshooting tool. It was quickly determined that simply taking metric data and making it look prettier than other people's pretty data had limited life and the customers were demanding more.

That's where the relationship between Vizioncore and its parent company Quest really started to shine. I was actually still the Director of R&D of Vizioncore at the time and it was determined that it didn't make sense for us to halt feature development of vCharter and spend the massive amount of effort necessary to take it from the troubleshooting tool it was to the monitoring framework it needed to be. Quest had just launched a 5th generation Application Monitoring and Management solution called Foglight. The management team had seen a few demos of what Foglight was capable of and it was decided that we could easily make the best damn virtualization monitoring solution there is by leveraging the Foglight application framework. A joint effort between Vizioncore and Quest was forged to create this solution.

We generated a list of required features that was collected from more sources than I can count. Customers, analysts, competitors, community forums; just about anything that helped us determine what people were demanding from a monitoring solution. After the requirements were finalized it was time to do the fun stuff. The primary role that I played in the development of the product was engineering the rules and alarm details required from a virtual infrastructure perspective. I also had a secondary role in helping design the graphical presentation and layout of all the data being collected.

My objective, as it was explained, to me was simple (oozing with sarcasm); take my brain and what I know about virtualization best practices and deisgn and, turn it into analytical intelligence for the application. The end result took the combined efforts of large teams at 3 development labs (Toronto, St. Louis, and Chicago), 18 Months, and followed me through 3 different jobs within the Quest organization. Every person who played a role in the development of this product deserves the high praise that will come in the next quarter as vCharter Pro and Foglight for VMware become available.

At this point you may be asking "What's the difference between vCharter Pro and Foglight for VMware?" If you are simply looking at it from a virtualization perspective, the answer is simple...pretty much nothing. Vizioncore's vCharter Pro Quest's Foglight for VMware will have the same capabilities surrounding virtualization monitoring. The main difference comes when you want to extend your monitoring outside of virtualization and start looking at end-to-end user , Java, Web, and Database monitoring alongside your virtual infrastructure. Foglight has had a unique play in the application space for some time, and as virtualization starts to propogate within the large organizations, there is a need to relate the application performance with the details of the virtual infrastructure.

OK, enough history and differentiating between two products doing the same thing. What is it that is so cool and unique about this thing? First and foremost, which is what grabs everyone's attention, is that it is graphically stunning. The dashboards are laid out in a way that it is always easy to view different aspects of the virtual infrastructure and everything has a logical flow of going from view to view. By simply looking at the screen it is easy to almost immediately determine when an issue has occured and what it's criticality is. The components move and change speed based off how hard a particular metric is working.

That takes us to the next point; vCharter Pro is SMART. Through all the information that was collected from all of our sources, we were able to build a strong database of common issues and best practices that have been developed, preached and used across many of the largest adopters of virtualization. The end result was a set of rules and alarms that not only told you what was wrong, but why it is wrong, and will make the necessary recommondations on how to resolve your issues. It gets better. it won't just tell you when things are bad...it also lets you know when things are too good! I'm sure I'll get a higher than 50% response if I ask "Who out there has run into the application owner than demands his application get 2GB of memory because it is so important to the performance of the system, and there is no way to show him that he is wrong?" Well, vCharter Pro can provide the necessary data to hand a report to the application owner and be able to confidently say "I'm taking you down to 1GB". If that wasn't enough, it also lets you know when something is flat out misconfigured; which beta testing has shown, happens A LOT more than I would have thought.

vCharter Pro is sprinkled with the best parts of the original vCharter, and is combined with a solid and scalable framework. In fact, our best GUESS (since we can't build an environment large enough to break scalability nor will we commit to an exact number) is that we can monitor a 10,000 virtual machine environment with a single management server, which can collect data from multiple VirtualCenter instances. The final thing that people are getting excited over is that every dashboard, every rule, every alarm, and every SLA is completely customizable. If you don't like a view or you want to add more data, chagne it. If you have a specific SLA that states you must have 20% overhead capacity on each ESX host for failover, create it. It's all within the end user's power to modify just about anything they can see or do with the data.

Trial downloads should be available from Vizioncore in Mid-April, but for now I want to thank those that took the time to actually read this horrible piece of shameless marketing and hopefully allow them to enjoy some of the screenshots and start to see the possibilities and cost savings in support and training that a product like this can provide.