VMGuru.nl – I choose (a virtual) life!

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.

(more…)

Last week a colleague, who sells applications running on an Oracle Database, had some questions regarding Oracle and running it in a Virtual Machine (VM) on top of a VMware infrastructure with a customer.

1) How to license Oracle in a virtual environment?

I pointed him to an article about licensing the Oracle software in a virtual environment I wrote some time ago.

Oracle can namely be hard- and soft partitioned, where VMware, XenServer, Hyper V and Oracle VM are all marked as soft partitioning, while looking into the way Oracle VM can be hard partitioned I stumbled on the following how to do it:

There are two methods to pin virtual CPUs. You can use the xm command to pin a guests’s virtual CPUs or you can hardcode the CPU mapping in a guest’s vm.cfg file. The difference between pinning CPUs with xm and hard coding the CPU mapping in a guest’s vm.cfg file is the persistence of the CPU mapping. CPUs that are pinned with xm are not persistent between reboots. Hard coding the CPU mapping in a guest’s vm.cfg file is persistent between reboots. To comply with Oracle’s hard partitioning policy, you must hardcode the CPU mapping in a guest’s vm.cfg file.

(more…)

About two weeks ago I attended a session at the VMware User Group meeting here in the Netherlands about Project VRC.  After the presentation I asked myself: ‘What is the value of this project?‘.

For you who don’t know what Project VRC is:

“Project Virtual Reality Check (VRC) is a joint venture of Log•in Consultants and PQR, who have researched the optimal configuration for the different available hypervisors (hardware virtualization layers). The project arises from the growing demand for a founded advice on how to virtualise Terminal Server and Virtual Desktop (VDI) workloads. Through a number of researches, Log•in Consultants and PQR show you the scaling possibilities for Terminal Server environments as well as Virtual Desktops.” http://www.virtualrealitycheck.net/

Don’t get me wrong: What they did was a very good initiative, it showed the performance differences between different hypervisors. Although the results were not that surprising it was good to see the validation numbers of the things we already knew.

I also think that the guys who did the project where totally surprised by the attention vendors and customers gave to the project. It was an outstanding (marketing) tool to show the value of virtualization and especially XenApp on a hypervisor. Because of this attention the whole project got out of hand. Although this was not the goal of the project, vendors and customers used it as a reference guide for vitalizing XenApp. That’s the point where I started to wonder what the real value of the project VRC was.

(more…)

At the end of May of this year we wrote a article concerning Hypervisor comparisons and we got a lot of positive feedback on it. The downside to that is that people want an update as soon as one of the companies launches a new version of its product, and who can blame them. However the issue is that this takes a lot of research and because of that, a lot of time. And because two of us are ill and in bed wearing a sombrero and the other two are extremely busy, we simply don’t have that time right now.

However, Scott Lowe has written an excellent article on the feature comparison between VMware vSphere 4 and Microsoft’s Hyper-V R2 which is a must read for everybody who’s advising customers on hypervisors.

It’s not as extensive as the Enterprise hypervisor comparison we did earlier but it gives you a good image how both products relate to each other. To extend the picture I added a list of supported operating systems.

(more…)

Project Virtual Reality Check finally posted a new document about previous results and possible clock drift when using the “Login Virtual Session Indexer (VSI)”.  Previous test setups and results didn’t take into account how different hypervisors handle passing time.

In my opinion this is a serious setback to Project VRC which is considered an institute in the virtualization world. People will start questioning the results when no new tests will be performed.

Below is a description from the Project VRC website explaining the new whitepaper they published on September 14th 2009. This is a must read for people that already did some testing as well as new tests. In short: ‘Because of Windows clock behavior in virtual machines the results were affected and some hypervisors may come out better than they really are.

This whitepaper is a review and reflection on previous Project VRC publications, the benchmark: “Login Virtual Session Indexer (VSI)” and Windows clock behavior within virtual machines.  This discussion is fueled by the fact that results from the individual Project VRC whitepapers are set side-by-side to compare hypervisors. Project VRC has been in discussion with both vendors and community, and performed additional research in this context. Before Project VRC can publish new results, it is important to address any questions, review the impact of this discussion and improve VSI where possible.

You can download it at www.projectvrc.nl

The major conclusions in this Whitepaper are:

(more…)

I was checking out the news and was checking out the competition on Microsoft’s Virtualization Team Blog and ran into some interesting news.

Microsoft just announced that their Windows Server 2008 R2 and Microsoft hyper-V Server 2008 R2 have been released to manufacturing.

I guess my ‘Microsoft-informant’ was right, Microsoft’s intention is to release Windows Server 2008 R2 and Hyper-V at the same time and not 3-6 months later as they did with version 1.0.

After the initial Hyper-V R1 release, Microsoft interviewed their customers to discover what they wanted in their new v2.0 product. These are the improvements and priorities which their customers want (according to Microsoft).

(more…)

I use TWeetDeck on my laptop to follow the news concerning virtualization. vSphere, Hyper-V and Xen have their own column. It’s because of these queries I noticed a tweet from @ilijabrajkovic about a free Hyper-V training from Microsoft.

(more…)

Last night, after my son’s first birthday,  I checked my missed twitter messages and a retweet from Duncap Epping caught my eye. It was a link to an article from Eric Gray on vCritical about HA for Linux guests on Hyper-V. Now you will probably think, nice an HA for Linux on Hyper-V whitepaper/how-to, but this is not the case.

As you probably know, very little Linux distributions are supported on Hyper-V and therefore no integration tools are available. Because the lack of integration tools Microsoft Clustering Services, the service which provides HA services for Hyper-V, you can not gracefully shutdown an unsupported Linux host and when shutdown MSCS will desperately try to keep the guest up and running. Because of this shortcoming you will have to choose, run HA OR Linux because you can not run both.

A complete desciption can be found here in Eric Gray’s article on vCritical.com. If you are considering a Hyper-V implementation you should definitaly read this article.

Yesterday I attended the ‘TechNet – Get Ready for all Microsoft virtualization products’ session in Utrecht (NL). This was a new style paid (€99,-) Technet session, max of 20-30 professionals in a classroom. The program consisted of Hyper-V, App-V, Med-V, Terminal services and System Center Virtual Machine Manager 2008.

Hyper-V
We started with an introduction to Hyper-V, luckily the trainer had updated his lab to Hyper-V 2, so we could check out the new features which should compete with VMware. After explaining the architecture, networking and storage it was very obvious that this was the standard Microsoft propaganda. They were comparing Hyper-V 2 to VI 3.5, which were practically the same, but of course Microsoft’s hyper-V solution was much cheaper. Grrrrr! Again, when you do a comparison, do a fair comparison and compare Hyper-V to vSphere 4 (because the release is not far away I will allow Microsoft to use Hyper-V 2 in this comparison). Then the numbers are very different and maybe the VMware solution costs more but obviously you will get a lot more.

I also disliked the insinuation that when they compare a microkernel hypervisor to a monolithic hypervisor Microsoft  makes it look like VMware’s hypervisor is one out of the stone ages. Play fair, win the fight based on your own strong points.

When he explained the Hyper-V I/O architecture with the parent partition, VM Bus, synthetic drivers, etc, I asked him if this didn’t introduce a singe point of failure and a possible I/O bottleneck. In true MS fashion he denied both, claiming that I/O was tested and you shouldn’t install Exchange in the parent partition. Duh!

(more…)

Browsing through Planet V12n I stumbled upon a great article by Steve Kaplan, which I didn’t want to keep from you.

As we all know we’re in the midst of a hypervisor battle. VMware vs (Microsoft / Citrix).

VMware focuses on their broad range of products for server-, desktop- and application virtualization including the necessary management and automation tools and disaster recovery solutions to enable the virtualization of a complete data center and desktops.

Microsoft focuses on the cost perspective and claims that their Hyper-V solution is cheaper then VMware’s vSphere.

No problems so far but if you want to compare cost, keep the fight clean and do an honest comparison. In his article Steve points out the areas were he thinks the comparison is not fair or not clear. I’ve read the comparison multiple times and I must agree this is one of those cases where Microsoft does not play fair as we encountered earlier with Citrix’s misleading comparison of HA, 2.

In my opinion a great job by Steve and a must read article.

April 25th we published our Enterprise Hypervisor comparison and we got very positive comments on it. A few people were kind enough to provide us feedback so we could improve the document, thanks for that.

I collected all comments and feedback and created version 1.3 of our Enterprise Hypervisor comparison which can be found here.

Again, feel free to contact us when you have feedback for us so we can improve the list.

(Gabrie, thanks for the detailed feedback. I hope  you will find that all points are taken care of.)

Update 4-6-09: Updated to version 1.3 after feedback from Jorge

As you may have noticed, we at VMGuru.nl have been extremely busy the last few week. Tons of work accompanied with the right amount of stress, illness and personally the build of my new house. This certainly shows in the amount of article we posted lately.

But I try to keep up-to-date, certainly with the release of vSphere 4. So today I read a great article on vTeardown by E. Horschman which I did not want to keep from you.

The article describes the quick migration feature in Microsoft’s Hyper-V, compares this to the live migration feature in the new version and warns customers on the  ‘Quick’ features provided in the next version of Hyper-V.

In the past Microsoft claimed that their Quick Migration feature was enough and Live Migration, as VMware’s vMotion, was not needed.  In vTeardown’s article you can see how quickly Microsoft changes its opinion as soon as they have finally added the feature themselves.

‘Now that Microsoft has live migration on their Hyper-V roadmap, we’re starting to learn that they never really thought much of Quick Migration themselves.  I heard Mark Russinovich, one of their technical luminaries who gave a talk about Hyper-V R2 futures, actually come out and say what we all knew to be true, “Quick Migration was our first attempt to do a live migration, and to put a nice spin on it, we called it Quick Migration. [...] Even though we said, [...] ‘trust us this is really cool, this is what you want, you don’t want instant, that’s not as good as this, this is quick,  but people didn’t seem to buy that, so we ended up [...] implementing live migration, so that Quick Migration stuff is crap, this is really good.’”  Mark’s honesty got some laughs from the packed session, but it puts us on notice that we should be skeptical when Microsoft tags a feature as “Quick”.

Not very trustworthy in my opinion. E. Horschman agrees and adds a warning to beware of Microsoft’s ‘Quick’ features as they now use the same b*lls*!t motivation on their new Quick Storage Migration feature.

So check out ‘When Microsoft Says “Quick”, Do They Really Mean “Crap”?‘.

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.

(more…)

Today my twitter popped up and I saw a retweet from Duncan Epping pointing to an article on VInternals. Duncan added a warning ‘Please sit down before reading my latest post‘, this caught my attention.

I read the article and I must say, this is a MUST READ for all virtualization architects, admins, decision makers!

The article, published January 2009, contains information directly from employees from Microsoft explaining the design considerations in relation to Hyper-V. Make sure you’re sitting down when you read the excerpts, are somewhere private like your own home, and have the bathroom door open and an Ambulance on hold. Because it is very likely you will piss yourself laughing.

This article is the next article in a long sequel of Microsoft Hyper-V vs VMware ESX articles, like:

• ESX vs Hyper-V mythbusting myth
• VMware releases a cost per application calculator
• Hyper-V, not in my datacenter!
• VMware ESX vs Microsoft Hyper-V

VMware dedicated a blog to this so called Microsoft mythbusting, you can find it here.

Last March I wrote an article on vendors  trying to market their products by bashing the competition and not having a clean discussion. At that moment I thought ‘Am I the only one making this into a problem?’  Today I read a great post by Gabrie van Zanten of Gabes Virtual World and it seems I am not alone!

Gabrie watched a video on Microsoft’s Virtualiztion Team Blog in which two gentlemen try to bust 10 VMware ESX vs MS Hyper-V myths. I intentionally use the word ‘try’ because they do a pretty terrible job.

They compare the current release of VMware ESX 3.5 to the R2 version of Microsoft’s Hyper-V which still has to be released (release 2010(?)). Now this is something that sounds familiar, this is exactly what I commented on in this article. Citrix comparing VMware and Citrix HA and not explaining that Citrix HA is Live Migration. If the MS gentlemen want to do some real mythbusting, try comparing VMware ESX 3.5 U4 to Microsoft Hyper-V 1.0 (both out now) or compare VMware ESX4/vSphere to Microsoft Hyper-V 2.0 (both not publically available yet).  But I think they’re too chicken to do that and they are right. Microsoft would come out way behind if you do a real comparison and claiming to have an enterprise ready hypervisor.

In my opinion in these kind of marketing campaigns you can recognize good vendors/product from bad ones. When you are not operating out of your own strengths and weaknesses but need to stoop down to bashing the competition to try to look good, you’re a myth yourself.

Want to see what all the fuss is about read the complete article in which Gabrie busts some mythbuting himself, look here!

So , Gabrie thanks, I’m not alone any more

Last few weeks I tested Hyper-V with the intention to write an article on the subject and give you a good idea how Hyper-V compares to VMware. After a few days I was not sure if I was objective enough and if this would be a good idea. To be honest I’m a big VMware fan and during the Hyper-V test I realized this was a true MS product. What that means, it’s buggy not consistent and has some strange MS ‘brainwaves’. So certainly not my product but maybe my view is colored. But luckily I’m not alone. Gabrie wrote some great articles on the subject and his opinion is ‘Hyper-V, not in my datacenter‘ (more…)

On Gabes Virtual World I read an article with the title ‘Unbelievable Hyper-V performance ‘. At first I was a bit shocked by the title. Hyper-V good performance? That the complete opposite to my own experience with Microsofts virtualization attempt. But luckily what Gabe ment was ‘Unbelievable BAD Hyper-V performance ‘.

You can read the complete article here.

Conclusion
A max score of 1,875 VMs per core where VMware ESX on the same hardware does 3,5 VMs per core. Maybe the test is not 100% accurate but with these scores maybe someone should step up and do a real hypervisor performance comparison (same hardware, load, etc).