Why are you monitoring your desktops? I thought VDI was all about your users and their applications. Xangati brings an industry first for VDI, monitoring based on End-User Experience (EUX)\User Profiles. The newest release of the Xangati Managemsent Dashboard (XMD) Suite 5.0 integrates with VMware and Citrix connection brokers. In Xangati like fashion this is all done agent-less. For a good little discussion on agent vs agentless check this discussion out. I agree with the Deltona on 2nd comment, agents should be avoided at all cost.
The User Profile Monitoring allows sanity, in the insanity that sometimes can result in the deployment of non-persistent/stateless desktops. As the user’s machines get destroyed after they get log out or get recomposed, their performance profile with remain. Having a problem on a desktop and then finding out about it 30 minutes later to find out there is no user on the desktop, or there is no desktop or the user is now in a different pool can make a person’s head spin.
The ability to have recorded events that gives visibility into all the layers of the stack really is powerful. You can visually see what the desktop is doing, what applications are running, what resources they are being consumed and how the display protocol is performing. Below are some graphs of a desktop. You can bring up the performance charts side by side and visually see the impact on one another. Pictures don’t really do it justice, to see it in full effect you really need a demo. The demo’s can be found at http://www.youtube.com/embed/Uk-Nw4oQ-58?rel=0
The Xangati Management Dashboard (XMD) along with the User Profiling added multiple vCenter support (Up to 8 vCenters) & best practice thresholds integrated from VMware. The multiple vCenter support is key to have a holistic VDI environment. You need to have the VDI Pods and Management Pods under the same view. If you don’t monitor the brokers and can’t log in, not much point monitoring the desktops!
I like the best practice thresholds because they are adjustable and I don’t have to configure it to get it going. XMD can take a up to two weeks profiling your environment before it will alert on problems that are out of normal range. The best practices can help point out errors during that time or if you decided to ignore an issue and it has now become the “norm”.
XMD also released Capacity Management into the Suite. I haven’t played around with it much but it looks like ok. I would give it any awards but all the information is there. It’s text heavy but your alerts are sitting there. I think they need capacity management to compete with the “check-lists” of other products. If Xangati can get IOPS and Network capacity management added it would be a big plus. Teaming up with Login VSI to keep the max thresholds for IOPS and network would be great for enterprise deployments. Xangati does have release schedule for the Capacity Managemen:
Phase I (Now)
Classic capacity trend management
Be forewarned about hitting capacity ceiling across all data centers
Find spare capacity/capacity hogs
Phase II (2nd Half 2012)
Live problem analytics informing capacity planning
It past releases there was the annoyance of a manual process to update the VDI dashboard with Clients and VDI Desktops. This is all automatic now, you can configure rules to include what you want to be added to the dashboard. Great to see this huge housekeeping task off the books.
The update will go a long way in VDI environment. I think the proof for its real time ability is the fact the VMware is running it it’s R + D. I’ve seen Xangati reports a couple time in some recent whitepapers.
The UI is coming along but the meat and potatoes are being served, come get it!