Back by popular demand, the Enterprise Hypervisor feature comparison.
After the release of our latest comparison I’ve received a lot of requests to include RedHat’s RHEV to the comparison. Although I’ve never encountered it in enterprise environments, I decided to add it as a service to our readers.
I based the RedHat features on their 3.1 version which is in beta right now. This is because I’ve limited knowledge of the product and I received an updated comparison from one of our readers based on this version.
I hope you find the new Enterprise Hypervisor comparison useful and feel free to contact us when you have feedback for us to improve the list.
Two weeks ago VMware released the new version of their vSphere hypervisor, so it’s time to update our Enterprise Hypervisor comparison. It very impressive to see how quick VMware has reacted to the Hyper-V 3 announcements and has taken most of the wind out of the Microsoft sails.
I hope you find the new Enterprise Hypervisor comparison useful and feel free to contact us when you have feedback for us to improve the list.
The information on Microsoft Server 2012 Hyper-V features is very inconsistent, many different values out there.
In this version I added 10 new criteria. Many of these criteria should, in my opinion, be available in hypervisors suitable for enterprise environments.
During the last few years we published several Enterprise Hypervisor comparisons and we got very positive comments and feedback on it. With the release of vSphere 5, XenServer 6 and a service pack for Hyper-V it was time for an update.
It very interesting to see how some of the products have improved over the years and how the three major manufacturers look at each other and copy features. But you can’t trust all manufacturers by just a simple green checkbox. Some claimed features need third party add-ons, aren’t suitable for production workloads or are only supported on a limited set of operating systems. You have to investigate further and I hope I’ve done most of that work for you with this new enterprise hypervisor comparison.
We here at VMGuru.nl hope you all had a proper start of 2011. We sure did!
But what will 2011 bring us? Some curtains have been lifted a little. We know we can expect a brand new version of vSphere in 2011. But what features will vSphere 5 have. Personally I hope we will finally see a vSphere Client for Linux/OSX with this version.
We will certainly see Citrix and Microsoft push the boundaries more and more in the desktop virtualization area. But do you think we will finally see the major shift to 64 bit in this field too? It has been predicted for years and years (some folks even mention 128 bit systems in the near future).
One of my coworkers pointed me to a video from this years Tech-Ed Europe about VDI protocols. Bernhard Tritsch did an interesting comparison between the different remote protocols used in today’s VDI solutions. In a 60 minute session Bernhard explains the differences between location (host vs client), type (hardware vs software) rendering and compression types (lossless vs lossy).
Although the results aren’t that good for PCoIP (software version) it still is a very interesting video.
Last year we published an Enterprise Hypervisor comparison and we got very positive comments and feedback on it.
During the last few weeks I received many update requests so I decided to update the old hypervisor comparison but this time I changed the setup a bit.
Changes:
No beta or pre-release versions are used. In the last document we also compared Hyper-V R2 beta which wasn’t officially released.
This time all software is available and no features are subject to change due to beta-test, etc.;
The versions used are the platinum/ultimate/fully-featured versions of the hypervisors. Product features can be limited by lower license versions;
No free versions have been used in this comparison.
It is always nice to see how Citrix fan-boys reacts on publications about VMware View. Especially when it comes to an comparison or XenApp for View trade in. It is often said that Citrix listens to their customers but they are seldom heard! People need simple solutions in a complex IT world. Why pay for a very high end piece of engineering like a Ferrari if I can get the job done with a Volkswagen! OK, it is nice to have a car that nobody else has and can do 320 km/h but in times were budgets are under pressure, I have a hard time defending expensive, complex and high-end solutions.
The fact that VDI is putting a significant load on the storage is true, but that counts for every VDI solutions including XenDesktop, so no differences there. Also the new upcoming version of View includes a basic profile management solution “Persona Management”. Which makes the use of linked clones a very workable solution.
When you compare installing View and XenDesktop you will see the following diagram.
When it comes down to management consoles the two solutions also look very different.
After attending the dutch Citrix Partner Exchange 2010 I realized that there is a lot of FUD out there:
in the Citrix community with regards to VMware View and PCoIP;
with me personally with regards to XenDesktop (no F, but a lot of UD)
This is also what we saw when Alex shared his experience with Citrix XenDesktop, which was not so positive, and we got a lot of comments comparing XenDesktop to VMware View.
But the Citrix Partner Exchange got me interested in XenDesktop and XenClient and I decided to do a little research. Then I came across Brian Maddens site to find that he had just finished his ‘Geek week VDI‘ in which he did a ‘real life’ lab-test with VMware, Citrix and Microsoft VDI. They tested all three vendor in their lab environment but added a WAN ‘simulator’ to create real life and worst case scenarios by introducing packet loss and latency.
And honestly I was surprised by some of their their conclusions. Not because I have no faith in Brian Madden but because I know Brian Madden to be a real Citrix enthusiast and a PCoIP critic. At the end he was very honest by admitting that Citrix XenDesktop looks like a mash-up of a bunch of different things, he was surprised by the simple, straightforward installation and configuration of VMware View and the good performance of PC-o-IP.
So bottom line?
VMware View shines because of simplicity and has good user experience even with PC-o-IP over a WAN connection.
XenDesktop is, at the moment, certainly the more mature and complete product but it’s complexity is a drawback.
And Microsoft ‘in box’ VDI? Well as expected, it’s complex, not enterprise ready and it’s no match for Citrix or VMware.
So as always, there is no clear winner, it all depends on the customer’s wishes.
Special thanks to Brian Madden and his team for creating this great VDI test!
I won’t summarize the total VDI test, you can read it yourself here but I will quote some of the conclusions which I found to be very interesting.
Based on the real life results when virtualizing XenApp I thought it was about time to summarize some of the best practices for virtualizing XenApp servers.
Why we DO want to virtualize XenApp?
For server consolidation: vSphere enables scale up XenApp deployments;
For mixing server editions: 32-bit and 64-bit XenApp VMs can coexist;
For management: Better management through flexibility & isolation think about Change Management and VMware DRS;
For high availability and disaster recovery: VMware HA and vCenter Site Recovery Manager;
For less costs for server hardware, maintenance contracts, power, cooling, floor and rackspace.
Virtualizing XenApp servers is very complex. There are a lot more layers involved, like the type of hardware, the capabilities of the processor, the performance of the shared storage, the hypervisor used, the specific settings per hypervisor, operating system settings in a virtual environment, the XenApp settings in a virtual environment, the Workspace management settings in a virtual environment etc, etc.
In the following sections I tried to summarize some of the best practices we use in our projects:
Last week I held a presentation for our Sales department. As we sold a lot of Citrix Xantippe over the last few years they wanted me to tell them more about the differences between a VMware View and the traditional terminal server based Citrix XenApp solution. So I made a summery about the differences and similarities. When VMware also asked me about the presentation I decided to translate it into English and share it with you all.
Yesterday Citrix announced HDX 3D which is an enhancement to the ICA protocol which, according to Citrix, delivers 3D applications by only using one-tenth of the bandwidth.
This is fierce competition for VMware’s PCoIP which they developed with Teradici and which VMware will demo at VMworld 2009 this week.
I have only seen the HDX 3D demo video, which can be found here, and I’m anxiously waiting for footage of the PCoIP demo at VMworld 2009 this week. I’ve seen the demo at VMworld Europe 2009 and I must say I was very impressed . On the other hand I found the Citrix videos to be very shaky/flashy/blinky. I hope (and to be honest, I think) that this is not representative of real life performance (I’m in a pro Citrix mood today ).
If you want to know what HDX 3D is and how it works please visit BrianMadden.com. He can explain this much better than I can but you will have to see past the Citrix minded VMware skepticism ‘If VMware can actually come out with a decent implementation of PC-over-IP’-like comments.
Burton Group is in their last day of their annual Catalyst Conference in San Diego (July 27 – 31, 2009). Wednesday they had their announced VMware vs Citrix battle and I just watched their video feed.
It was a very interesting ‘battle’ between VMware’s Scott Drummonds and Citrix’s Simon Crosby but I must admit that I was a bit annoyed a times. I think the the photo on the left captures the atmosphere for the debate quite well.
Simon Crosby constantly attacking VMware on various issues, but mainly on VMware’s standpoint on performance testing, and Scott Drummonds staying very relaxed, not getting into a bitch fight, throwing mud and pulling hair.
As I am in the final week before my holiday and all projects are finished, I thought it might be a good idea to check out the competition. So I downloaded the free version of Citrix’s XenServer 5.5 to see what the fuss is all about.
First I tried to install it on a spare desktop at home (P4 2,4GHz CPU, 2GB RAM) but this failed, probably because the CPU doesn’t supply Intel VT.
Yesterday at work I grabbed two desktops with a Core2Duo E6400 with Intel VT and 4GB of RAM and installed Citrix XenServer 5.5 on it.
The installation is pretty straightforward, much like ESXi, and in about 5-10 minutes you will have a XenServer up and running. The first thing I noticed is that the console is very feature rich. We all know the yellow/gray console of an ESXi server and all you can do is basic configuration tasks like set an IP address, DNS server and default gateway, give the server its name, do a few tests and that’s about it. The console of a XenServer has a lot more features. Next to all ESXi-like features you can start and stop virtual machines, configure storage, resource pools and licensing, configure back-up, restore and updates.
The last few weeks many blogs and forums have spend time on hypervisor comparisons and I have read tons of articles on the subject. Many only compare hypervisors based on performance, features or cost. I think it’s a bit more complicated then that. After Citrix announced that their XenServer product is available for free I spend a fair deal of my time explaining to colleagues and clients that this is a hoax and that cost is not the only reason to base their decision on. Especially in the case of XenServer the choice and the long term effects make it a little bit more complicated.
When I read Chris Wolfe’s article on ‘Production-class Hypervisor Evaluation criteria‘ and saw his VMworld Europe 2009 presentation (DC15) I found someone who read my mind. Chris knows what he is talking about and uses the right criteria to select the right hypervisor for the job. Now you probably think ‘These VMGuru.nl guys are VMware fans so here we go again‘ but the opposite is true.
Like Chris I think every situation has its own ideal solution and you should select the hypervisor based on well-considered selection criteria and because my employer, Centric, focuses on clients with 500+ workstations/employees these criteria are Enterprise-class hypervisor selection criteria.
Now most of our loyal followers know, I usually do not post any content on this blog other than things related to the hosting of website. It’s my “job” to keep this part of the infrastructure running (well, that’s not quite how it is, but that’s another subject for another post ). In my day-to-day job I see a lot of infrastructures pass by and recently I started to notice that some of my colleagues look at virtualisation as an application. They think in boxes.. “we need a DC, and a filer, and blahblahblah”…So I thought, let’s post an entry here and see if you all have something to say about it.
Most of the time when I’m frustrated it helps to play some squash, ride my bike or vent my frustration on this blog. As all walls in the area have been smashed to bits due to frustrations in the past and it is too cold to ride my bike, I will try the last one.
Yesterday I attended a seminar from work together with Edwin, Anne Jan and Alex. The subject was virtualization with Citrix and we were invited to keep a good balance between Citrix and VMware but we could not give a presentation due to time constraints. The people who organized this event know us pretty well and know (and share) our love for VMware so we were asked not to mess up the presentation and bash the Citrix guy. And so we did………
Today we had a very interesting presentation from Citrix. At one point the discussion was about the Citrix’s definition of High Availability.
I really can imagine that this is getting a confusing discussion. VMware says: HA is to resolve a single host failure. Citrix however also calls live migration a form of HA.
I can’t argue with that but to keep the discussions and comparisons clear it is essential to specify what you define as HA. I think both are right, but the trouble is that, when comparing products, people think they have a solution for hardware failure, but instead they can do a live migration.
I hopefully that in the future both parties start using the same names for the same options, or at least explain what their option is.
Back by popular demand, the Enterprise Hypervisor feature comparison.
Two weeks ago VMware released the new version of their vSphere hypervisor, so it’s time to update our Enterprise Hypervisor comparison. It very impressive to see how quick VMware has reacted to the Hyper-V 3 announcements and has taken most of the wind out of the Microsoft sails.
During the last few years we published several Enterprise Hypervisor comparisons and we got very positive comments and feedback on it. With the release of vSphere 5, XenServer 6 and a service pack for Hyper-V it was time for an update.
Yesterday, the three of us attended one of the quarterly 
).
It was a very interesting ‘battle’ between VMware’s Scott Drummonds and Citrix’s Simon Crosby but I must admit that I was a bit annoyed a times. I think the the photo on the left captures the atmosphere for the debate quite well.
). In my day-to-day job I see a lot of infrastructures pass by and recently I started to notice that some of my colleagues look at virtualisation as an application. They think in boxes.. “we need a DC, and a filer, and blahblahblah”…So I thought, let’s post an entry here and see if you all have something to say about it.
Most of the time when I’m frustrated it helps to play some squash, ride my bike or vent my frustration on this blog. As all walls in the area have been smashed to bits due to frustrations in the past and it is too cold to ride my bike, I will try the last one.
