VMotion

You are currently browsing articles tagged VMotion.

Virtualization Short Take #24

Monday, December 15, 2008 in General by slowe | 1 comment

There’s lots of good information flowing around the Internet, and it’s becoming increasingly difficult to sort through all the useless stuff to find the valuable gems. Hopefully, some of the links that I have collected here will prove to be more useful than useless!

  • I came across this VMware KB article titled “Dedicating specific NICs to portgroups while maintaining NIC teaming and failover for the vSwitch”. I was hoping it would shed new light on some NIC teaming functionality. Unfortunately, it was only about overriding the default vSwitch failover policy on a per-portgroup basis. I was already well aware of that functionality and use it quite extensively in my VMware designs, but for others that may prove useful.
  • This video about VMware DPM sparked some debate about spin-up/spin-down affecting drive MTBF and decreasing a VMware ESX server’s operational lifecycle. Chad Sakac of EMC shared some findings from EMC regarding spin-up/spin-down in this post and came to the conclusion that using VMware DPM should not materially affect the reliability or lifetime of servers (at least with regards to drive failures). Personally, I tend to agree that this was FUD, most likely from a competitor, but it’s best to get this sort of thing out in the open and debunked.
  • Leo posted a brief snippet of code to upgrade the VMware Tools on VMs without a reboot. It looks like it might come in handy. And Leo’s guide to configuring jumbo frames with an EMC AX4-5i is quite useful, too—it’s a nice counterpoint to my own guide to configuring jumbo frames.
  • Tomas ten Dam has completed his guide to building a complete “SRM in a Box” setup using the NetApp Data ONTAP Simulator. Of course, Chad wants him to use the Celerra VM…
  • Oh, and while we’re talking VMware SRM, be sure to check out Mike Laverick’s book on VMware SRM, “Administering VMware Site Recovery Manager 1.0″. I haven’t read the book yet, but knowing Mike I’m sure it’s good quality stuff. Maybe Santa will give me a copy for Christmas.
  • Sven H. over at VirtualFuture.info posted a good guide on using thin provisioned VMDKs with VMware ESX 3.5 via the vmkfstools command. (I was going to include a trackback to Sven’s post, but his blog theme doesn’t show the trackback URL.) Seems like I saw somewhere that thin provisioned VMDKs in ESX 3.5 are still unsupported, so deploy accordingly.
  • Via Tony Soper, I found that version 2 of Microsoft’s Offline Virtual Machine Servicing Tool is available. I first discussed the Offline Virtual Machine Servicing Tool back in June during Tech-Ed 2008. You can download the tool here.
  • Also from Tony, here’s a great article on how to balance VM I/O with Hyper-V. An interesting tidbit from this: by default, I/O balancing is enabled for storage, but not for networking. I can see it needing to be enabled for storage, but why disabled by default for networking?
  • More information on controlling resource utilization within Hyper-V is provided in this article by Robert Larson. It’s worth having a quick look if you are unsure how to configure it or how it works.
  • Ben Armstrong answers the question, “Why does it take so long to create a fixed size virtual hard disk?” The answer: the disk space is zeroed out in advance. My question is this: is this need to zero out the disk space a result of how NTFS deletes files or is this scenario applicable to VMFS as well?
  • This has probably been mentioned before, but users considering virtualizing their Active Directory domain controllers should keep these considerations in mind.
  • I recently ran into a situation where we need to change the IP address of an NFS datastore. (It’s a long story as to how this came about.) In any case, I told the customer that I couldn’t be sure that changing the IP address wouldn’t cause problems. Fortunately, before the customer tried it, I found this post by Rick Scherer. The short story: it doesn’t work, and you shouldn’t do it. Create a new datastore with the correct IP address and use Storage VMotion instead.
  • For even more information on Storage VMotion, also check out Chad’s post here.
  • VMwarewolf continues his Resolution Path series with common fault issues in VMware Infrastructure. Good stuff.

It’s clearly been too long since I published one of these, as I still have other links collecting dust in my “link bin”:

Third Brigade offers free security for up to 100 virtual machines
Version 4 of the PowerVDI tool
Go Daddy Wildcard Certificate with VI3
New VMware VI network port diagram request for comments
Auditing ESX root logins with email…

Like I said, there’s just so much information! And now that I’m trying to delve deeper into the storage realm, that’s only doubled up on the information I’m trying to manage. Hopefully I’ve picked out a few gems for you this week. Thanks for reading!

Tags: , , , , , , , , ,

Virtualization Short Take #23

Wednesday, November 19, 2008 in Virtualization by slowe | No comments

A short while ago, I had a colleague in the blogging industry ask me if the reason I was writing “Short Takes” was because I was too busy to write in-depth articles. At the time, I told this colleague no, but now I’m wondering if I should change that answer…

In any event, here’s my list of links and tidbits that I found interesting, amusing, or useful over the last week or so. Enjoy!

  • Eric Sloof of NTPRO.NL points out that VMware has updated their best practices to “allow” the placement of VM swap files on NFS, rather than recommending VMFS. Does anyone have a link to the VMware KB article with those updated recommendations?
  • Edward Haletky points out a process for performing “secure P2V” operations. Because the P2V process generally involves network communications with VirtualCenter and/or the VMware ESX Service Console, a straight P2V process would cross security boundaries. Edward’s phased approach helps with that issue.
  • Into PowerShell, but finding the process of modifying Offload Policies too difficult? This should help out significantly.
  • Expanding upon some earlier work, Rick has added “Resolution Paths” to common network issues and common licensing issues. Excellent work!
  • Leo provides some good information on monitoring VI3 with Zenoss.
  • Via Alessandro and Ben, I learned about the HVRemote tool for automating the configuration steps for enabling remote management of Hyper-V. HVRemote was created by John Howard and more information is available here.
  • Symbolik shares his experience in troubleshooting a PSoD (Purple Screen of Death) with VMware ESX. In his case, the issue turned out to be related to NICs, but it’s nice to see that he was able to zero in on the issue and get it resolved.
  • Larger installations with lots of LUNs and an active/active SAN may find this script published by Duncan very helpful. The problem that this script addressed underscores the need for robust multipathing support in VMware ESX such as that which will be allowed via the new vStorage APIs.
  • Need more information on Enhanced VMotion Compatibility (EVC)? Both Gabe and Rich have recently tackled the issue on their own blogs. If you’re wondering what EVC is, the short answer is that it’s a way to automate this process in a supported fashion.

That’s it for this time around. Thanks for reading!

UPDATE: Eric Sloof responded about the updated recommendations regarding VM swap files. The new information was disclosed in session TA2784 at VMworld 2008. See slide 38. Thanks for the information, Eric!

Tags: , , , , , ,

VMware Keynote – Automating DR with VMware SRM

Thursday, September 18, 2008 in Virtualization by slowe | 1 comment

There is no general session this morning at VMworld 2008; instead, a “keynote” will be delivered about automating disaster recovery (DR) using VMware Site Recovery Manager (SRM). This is similar to the way in which other vendors have delivered various “keynotes” throughout the conference instead of all the announcements being crammed into the morning general sessions.

The speaker this morning is Jay Judkowitz, the product manager for VMware SRM. I’ve met Jay before; he’s a good guy. There’s a small technical glitch as the session begins because the slide deck doesn’t come up, but that gets resolved within only a few minutes and Jay begins his presentation.

The presentation begins with yet another overview of the VDC-OS vision; SRM is considered one of the vCenter management vServices. Jay then goes on to address all the various ways in which VMware provides application availability for applications hosted on VMware Infrastructure. This would be technologies like VMotion, VMware HA, VMware DRS, VMware FT, NIC teaming, storage multipathing, and of course Site Recovery Manager.

The traditional challenges of DR (including complex recovery processes and procedures, hardware dependence, inability to test extensively or repeatedly) are all addressed by VMware SRM. More accurately, they are addressed by the products that form a foundation underneath VMware SRM. Features like hardware independence, encapsulation, partitioning and consolidation, and resource pooling. These features have a direct play in a DR environment. It’s funny to see Jay taking this particular approach; it’s almost like he’s using the same slide deck that I’ve used in DR presentations given over the last couple of months.

That finally brings the discussion around to Site Recovery Manager specifically. Jay goes over some of the features of SRM, and discusses some “do’s and dont’s” for SRM. For example, SRM isn’t really intended to provide failover for a single VM, although you can architect it to do that (put that VM on a single LUN by itself and create a Protection Group for that LUN and VM, then craft your Recovery Plan).

It’s important to note that SRM is not a replication product, but instead relies upon replication products from supported partners. This is done via the Storage Replication Adapter (SRA), a piece of software written by the storage vendor.

When setting up SRM, there are number of steps that it goes through. First, you have to integrate with the storage replication in place already (and yes, the storage replication needs to be in place already). Next, you need to map recovery resources; this creates the link between resources used in the Protected Site to resources that will be used in the Recovery Site. Third, you need to create Recovery Plans, which is the automated equivalent of the DR runbook. That is, the Recovery Plan defines which VMs will failover, in which order, at the Recovery Site. That’s a bit of simplistic overview but it does get the point across.

At this point, I’ve decided that I’m going to try to get into a different session. I’m quite familiar with SRM, a lot of readers are probably familiar with it as well, and it doesn’t look like there is anything new that will be revealed here. For those readers that aren’t familiar with SRM, let me know in the comments. If there’s enough interest, I’ll write something separate after my return from VMworld 2008.

Tags: , , , , , ,

BC2621: Fault-Tolerant VMs in VI: Operations and Best Practices

Tuesday, September 16, 2008 in Virtualization by slowe | 9 comments

Well, it looks like wireless coverage pretty much stinks, so this will have to be published post-session as well.

This session is BC2621, titled “Fault Tolerant VMs in VMware Infrastructure: Operations and Best Practices”, presented by Dan Scales, Principal Engineer with VMware. It’s a pretty full session, but I’m lucky (unlucky?) enough to get a seat toward the front of the session.

This is a technical preview of VMware FT, a new feature that is slated to be included in VDC-OS. This is the evolution of “Continuous Availability,” which was demo’ed by Mendel Rosenblum in San Francisco at VMworld 2007. VMware FT is part of the availability application vServices, with also include such things as VMware HA, VMotion, Storage VMotion, NIC teaming, and storage multipathing.

VMware FT is one of two new application vServices that are being discussed this week; the other is vCenter Data Recovery, a full backup solution.

VMware FT is based on vLockstep technology that keeps a primary and secondary machine in virtual lockstep with no special hardware. Like the rest of VMware Infrastructure, VMware FT will run on standard, x86 commodity hardware, provides zero downtime and zero data loss. This is just another level of availability that can be provided. When compared with hardware-based availability, which does not run on commodity hardware, or standard clustering solutions, VMware FT allows for more protection with less complexity at a lower cost.

Again, VMware FT is simply a way to provide a higher level of protection than VMware HA for some VMs and workloads.

VMware FT involves two VMs: a primary VM and a secondary VM. The secondary VM is doing exactly the same thing as the primary, but does not communicate across the network. (What about storage?) In the event of a hardware failure, the secondary VM will become the primary VM and will assume network connections. Like VMotion, there should not be any interruption in network connectivity. After a host failure, a new host will be selected to run a new secondary VM, and there is a brief window in which the VM can’t be fully protected with VMware FT.

VMware FT is fully integrated with VMotion and VMware DRS. Multiple FT pairs can run on a single host, and a host can run both FT-enabled and non-FT-enabled VMs on the same host. FT can be dynamically enabled or disabled dynamically, and VMware DRS is leveraged for the placement of the secondary VM.

VMware FT is based on VMware’s Record/Replay technology, which was first introduced in VMware Workstation in 2006. When the data stream is recorded, only non-deterministic events are recorded, and the replay will occur deterministically. This creates instruction-for-instruction, memory-for-memory identical results.

Deterministic means that a processor will execute the same instruction stream and will end up in the exact same state. By recording non-deterministic events, VMware ensures that the record and the reply are identical. Non-deterministic stuff involves things like network/disk/keyboard I/O and hardware interrupts.

Using record/replay, VMware FT keeps two VMs in lockstep. These two VMs will share a common disk, although in the future this may move to a non-shared disk. Only the primary VM responds across the network; the secondary VM is a silent partner. If the primary VM fails, the secondary VM takes over immediately. In the event of a secondary VM failure, then redundancy will be re-established by restarting the secondary and re-syncing the two VMs.

When using VMware FT, this must be done in a VMware HA cluster. Once in a VMware HA cluster, a VM may be protected by FT, HA, or both. When both HA and FT are in operation, then both technologies come into play. In the event of hardware failure, HA will restart VMs and FT will make secondary VMs take over and become primary while HA restarts the former primary VM.

When a user enables VMware FT for a VM, a special kind of VMotion is used to create a secondary VM on a second host (basically a copy of the configuration). Then the two VMs are kept in virtual lockstep via VMware FT. If the primary VM is powered off, the secondary VM is powered off as well; the secondary VM will also be powered off if VMware FT is disabled.

Now, looking at the hardware and software requirements, VMware FT requires CPUs that support hardware virtualization (AMD-V, Intel VT). These features sometimes need to be enabled in the BIOS of the server. All hosts must be running the same build of VMware ESX, shared storage is required (NAS or SAN), and all hosts must be in an HA-enabled cluster. In addition, a separate FT logging NIC and a separate VMotion NIC are required. This means a minimum of 4 NICs are necessary (Service Console, VM traffic, VMotion, FT logging). Gigabit Ethernet is required for the FT logging NIC (just like the VMotion NIC). There’s mention of “dedicated” or “separate” NICs; I wonder how firm that recommendation is? I’d be interested to know how this impacts best practices for NIC configurations.

VMware FT can’t protect VMs that are using thin provisioned disks; disks must be “thick.” Disks will be automatically made thick when VMware FT is enabled. (How does this impact vStorage Thin Provisioning?) VMs can’t have any non-replayable devices (USB, sounds, physical CD-ROM, physical floppy, physical-mode RDMs) and paravirtualization-based VMs are not supported. Otherwise, all VMs are supported, both 32-bit and 64-bit guest operating systems.

Once all the prerequisites are met, VMware FT can be enabled by simply right-clicking on a VM and selecting “Turn Fault Tolerance On”. After FT is fully ready to protect the VM, the icon color will change and the FT status will read “Protected”. A new “Fault Tolerance” pane within VirtualCenter will show all the VMware FT statistics and information, like the location of the secondary VM, the amount of secondary VM CPU and memory usage, latency, and log bandwidth.

The Fault Tolerance status will have a number of different states:

  • Enabled-Running (fully protected, this is the desired state)
  • Enabled-Starting (FT is getting started)
  • Enabled-Needs Secondary (a failure has occured, a new secondary needs to be created)
  • Disabled (VMware FT is disabled)

Disabling FT is better than turning off FT. When FT is disabled, all the underlying infrastructure is maintained and makes it easier to re-enable FT at a later date or time. Turning off FT removes all the underlying infrastructure and setup.

The secondary VM shows up as “VM Name (secondary)” in VirtualCenter. It will not show up in the inventory, but it will show up in the list of VMs for a cluster or in the list of VMs for the secondary host. This may be confusing but is based on customer feedback. There will be only certain places where the secondary VM will appear.

The Maps tab will have a way to show the link between the primary VM and the secondary VM.

There will be a number of FT-related events and alarms; the “Enabled-Needs Secondary” state, for example, is one place where an alarm already exists.

When considering VM migrations, either the primary or the secondary can be moved via VMotion, but both cannot be moved at the same time. There is a built-in rule in DRS to keep them on separate hosts. The DRS mode for fault-tolerant VMs is set to “Manual”; this means the user must explicitly choose to initiate a VMotion. FT must be temporarily disabled to do a Storage VMotion, or you could power off the VMs and do a datastore migration at that time.

Again, no network connections are lost during a failover.

Dan covered again the interplay between VMware HA and the placement of the secondary VM, and the use of a modified VMotion to provision the secondary VM. That is a very quick process, meaning that the FT status will return to “Enabled-Running” very quickly.

In the event of a multiple host failure, VMware HA will restart the primary and VMware FT will recreate the secondary VM to establish redundancy. In the event of a guest OS software failure, VMware FT won’t do anything because the primary and secondary are in sync; both will fail at the same time and in the same place. VMware HA failure monitoring can restart the primary; the secondary will then be recreated via special VMotion to re-establish redundancy.

What applications are suitable for VMware FT?

  • Applications that run well on uniprocessor VMs
  • Applications that can tolerate a small increase in network latency
  • Applications that have medium network bandwidth requirements (less than 600Mbps)

Examples of this would include medium-sized database applications, messaging applications, or important custom applications.

It’s important to keep in mind that the bandwidth of the FT logging NIC may become a bottleneck; watch how many FT pairs on placed on a single host or move to 10Gbps if available. You may also want to reconfigure some guest OSes (Linux with 1000Hz timer interrupts) to use a slower interrupt timer.

The session wrapped up with a customer portion by Mark Vaughn of The First American Corporation. Due to other schedule requirements, I didn’t stay for that portion of the session. Next up is some time in the Solutions Exchange. Stay tuned for more updates as soon as I can get network coverage.

Tags: , , , , ,

VMware’s Virtual Datacenter OS

Monday, September 15, 2008 in Virtualization by slowe | 10 comments

So, the news is out on VMware’s new announcement regarding the Virtual Datacenter OS (VDC-OS). You can find more coverage of this on a few other sites:

Virtual Datacenter OS from VMware
VMware moving to cloud-computing with Virtual Datacenter Operating System (VDC-OS)
VMware testing data center OS for managing, literally, everything
VMware Tries to Expand Throughout the Data Center

I’m kind of excited to be able to talk about this, because it was a topic of the Partner Technical Advisory Board today and I didn’t think I’d be able to talk about it until later in the week.

First off, let me just state that the Computerworld article linked above (“VMware testing data center OS for managing, literally, everything”) is just plain wrong. What VMware is working on is an OS for the virtual datacenter, not a virtual OS for the datacenter. The distinction is important. VDC-OS isn’t intended to be an OS for all aspects of the datacenter. It’s intended to be an OS for all aspects of the virtual datacenter.

When you think of an OS, you think of software that manages access to resources and provides services to applications. That’s what VMware is doing with VDC-OS: managing access to resources and providing services to applications, only this time the applications are workloads (virtual machines with an OS and a set of applications on that guest OS). VDC-OS will provide sets of services to these applications:

  • Application vServices, like availability, security, and scalability. These application vServices are provided via features like VMotion, Storage VMotion, VMware HA, VCB, and—in the future—stuff like VMware Fault Tolerance (FT), formerly known as Continuous Availability. See this page on VMware’s site for more examples.
  • Infrastructure vServices, like compute functionality (vCompute), networking connectivity (vNetwork), or storage features (vStorage). These vServices are manifested as features like VMware DRS, and will be extended in the future with things like vStorage Linked Clones, or 3rd party virtual switches, or VMDirectPath. The APIs are there for additional third party vServices to be added as well; one example would be network load balancing as an infrastructure vService.
  • Cloud vServices enable the interaction of on-premise infrastructure (the servers in your data center) to integrate with external cloud infrastructure. There are no concrete examples to really share here; in my opinion, this is the most nebulous part of this announcement. See this page for more information.
  • Finally, management vServices provide…well, management functionality for the virtual data center and the applications running in the virtual data center. More information is available here.

The way to really view VDC-OS is not as a “datacenter OS”, because it’s not intended to provide automation of non-virtual resources. Instead, look at VDC-OS as a framework. Within this framework are sets of services that can be extended or modified in very standardized ways (via APIs and SDKs) to provide different functionality for the applications running within that framework. Some third party ISV wants to write a different way of providing fault tolerance and failover? Fine, no problem, that can be plugged into the VDC-OS application vService framework for availability. I used the example earlier of a load balancer as an infrastructure vService. VMware announced the VMsafe APIs back at VMworld Europe, and the idea of VMsafe ties directly into API access for ISVs to develop new or different security-related application vServices that can be provided to all applications running within the VDC-OS, such as anti-virus or host-based intrustion detection/prevention.

I was a bit worried that this messaging wasn’t going to be received as clearly as I had hoped it would be, and the initial coverage I’m seeing so far confirms that many people are going to misread what VMware is trying to do. Hopefully, we can get the idea across to everyone so that they can begin to really understand where VMware is headed with this idea and why it is the next step in the evolution and maturation of virtualization.

Tags: , , , , , ,

Thinking on Microsoft’s Licensing Changes

Wednesday, August 20, 2008 in Microsoft, Virtualization by slowe | 2 comments

Much has transpired since yesterday, when I urged VMware to join the SVVP and get their software validated for full support by Microsoft. Since that time, it has come to light that VMware has joined the SVVP, although a formal announcement has not yet been made, and Microsoft has announced some significant licensing changes regarding virtualization. I’ve been reading the various announcements and analyses regarding this information and I thought it might be beneficial to try to pull all this together.

First, refer to Patrick O’Rourke’s blog entry, which does a great job of summarizing the need for application mobility licensing. Clearly, customers needed the ability to move applications freely between physical servers, and Microsoft themselves needed to allow customers to do this now that they have a more robust virtualization solution in place (Hyper-V and SCVMM 2008). While the licensing changes do benefit all virtualization vendors, it’s important to note that Microsoft needed these changes for themselves as well.

Patrick’s post also brings to light that while VMware has joined the SVVP, cooperative support is not yet in place. That won’t come until validation via SVVP is completed, which may take some time. The joining of SVVP was necessary, as it is merely one step toward a larger goal.

However, there’s more here than perhaps many people are realizing. Fortunately, there are a number of sites out there pointing out important caveats to the new licensing changes:

  • Rich at VM /ETC correctly points out that the new licensing does not apply to the Windows Server OS itself. So you are still going to have problems with VMware HA and VMware DRS automatically moving VMs from server to server unless you use Windows Server Datacenter Edition (see below).
  • Chris Wolf points out (both on his personal blog as well as the Data Center Strategies blog) that the lift on the 90-day license transfer does not apply to licenses purchased outside of a volume license agreement. Using OEM licenses? Then you’re out of luck; those licenses still fall under the old restrictions.
  • eWeek’s Joe Wilcox points out that because the Windows Server OS isn’t included in the 90-day license relief, some customers will simply license Windows Server Datacenter Edition for every CPU in their data center. Of course, the fact that you now get Hyper-V for free with Windows puts Microsoft…ahem, ahead of the game, shall we say? Read Joe’s full report here.

So, while Microsoft’s licensing changes are a good first step, there’s still more work to be done. Let’s applaud the changes, which were necessary, but let’s continue to press Microsoft to fix the issues that remain.

Tags: , , , , , , , ,

Virtualization Short Take #16

Monday, August 11, 2008 in Virtualization by slowe | 2 comments

There’s a lot of good information being shared out there! With all due credit to the original authors, here’s a few links that I found particularly interesting or useful. I hope you agree!

  • The VI Team Blog posted a list of interesting things in Update 2. They identify things like Enhanced VMotion Compatibility (EVC) and monitoring and availability enhancements. For example, did you know that VM failure monitoring is now fully supported and not experimental? One thing that caught my eye when I first announced Update 2 was VSS support. What I hadn’t really noted about this was the fact that this extends into the application layer, meaning that VSS-aware applications will be quiesced for VCB-based backups. This brings application data consistency to VCB-based backups, and this is a big deal.
  • And while we’re discussing the VI Team Blog, have a quick look at part 1 and part 2 of their tips for deploying VI. The discussion of Fibre Channel path selection policy with active/passive storage arrays is particularly helpful.
  • Rich over at VM /ETC brought some interesting facts to my attention regarding the free version of ESXi in this post. I was not aware, for example, that the Remote CLI was read-only with free ESXi. Very interesting, and quite useful.
  • Duncan at Yellow Bricks has also been pumping out some very helpful information, with an update on HA advanced options and a pointer to a document describing what happens if VirtualCenter crashes. One of the HA options that caught my eye was an option that allows VMotion interfaces to be used for HA. This means that we can easily provide additional HA redundancy simply by configuring VMotion interfaces. Handy!
  • Chris Wolf brings to light information on PXE booting ESXi, and in so doing discusses the idea of the “stateless hypervisor.” This is an idea that is gaining a lot of ground, also being the subject of the release of a utility designed specifically around that very idea (but for ESX, not ESXi). When I design a VI environment, I’m already treating the hypervisor as mostly stateless; all the data and “important” information is stored on the SAN. To fully embrace the idea of a stateless hypervisor, we also need to incorporate auto-configuration.

That’s it for now. If any readers have any interesting links they’d like to share, please do so in the comments below. Thanks!

Tags: , , , , , ,

Virtual Iron Tackles Power Usage with New Release

Monday, August 11, 2008 in Virtualization by slowe | 3 comments

Virtual Iron, the company I termed “the Rodney Dangerfield” of virtualization because their products just don’t get the attention and respect they probably deserve, is tackling power management and power utilization with a new release of Virtual Iron, version 4.4.

Version 4.4 adds a feature called LivePower, which builds upon Virtual Iron’s LiveMigration and LiveCapacity features to provide the ability to reduce power usage during times of low utilization. During periods of low utilization, virtual machines will be consolidated onto a smaller subset of servers—using LiveMigration so as to prevent any service interruption—and then power down the servers. When utilization begins to go back up again, the additional servers are powered back up and virtual machines are once again moved and rebalanced using LiveMigration and LiveCapacity.

The addition of this functionality brings Virtual Iron almost to complete feature par with industry leader VMware:

VMware VMotion = Virtual Iron LiveMigration
VMware Distributed Resource Scheduling (DRS) = Virtual Iron LiveCapacity
VMware High Availability (HA) = Virtual Iron LiveRecovery
VMware Distributed Power Management (DPM) = Virtual Iron LivePower

VMware still has the feature lead with other features like Storage VMotion, but Virtual Iron is a close second. In my mind, they are a more serious competitor to VMware than Citrix (Virtual Iron seems much more robust and mature than XenServer).

Full information on the new release and its new power management functionality can be found on Virtual Iron’s web site. The full press release is available here.

Tags: , ,

VMware Lab Manager 3.0 Released

Monday, August 4, 2008 in Virtualization by slowe | No comments

As has been already detailed elsewhere (such as by Alessandro or by David), VMware has released version 3.0 of Lab Manager. This lab automation software, primarily targeted at software development and QA environments, now brings greater integration with VirtualCenter. While I haven’t had the chance to fully review the release notes and the list of new features, my opinion is that the integration with VirtualCenter is huge.

Prior to this version, physical hosts running VMware ESX had to be configured for either VirtualCenter or Lab Manager, but not both. This also meant that Lab Manager couldn’t take advantage of VMotion, VMware DRS, VMware HA, etc., as all these functions were managed by or configured by VirtualCenter. With that barrier now removed, software development environments can now utilize these functions in conjunction with Lab Manager, and there is no longer a need to segregate VMware ESX hosts into separate farms based on whether they were being managed by Lab Manager or by VirtualCenter.

I would expect to see the development and release of products such as Lab Manager to accelerate. These kinds of products are “value adds” to the core virtualization offering, and VMware wants to be sure to offer customers a full virtualization solution, not just a best-of-breed hypervisor. Of course, that’s just my opinion, and I’ve certainly been wrong before.

Tags: , , , ,

More on Hyper-V for the ESX Engineer

Thursday, July 31, 2008 in Microsoft, Virtualization by slowe | 2 comments

My colleague and friend, Aaron Delp, recently published a post titled Hyper-V for the ESX Engineer. It’s a good post, and provides a good overview of Hyper-V for someone who might already be familiar with VMware Infrastructure 3 (VI3). With sincere apologies to Aaron for plagiarizing his title, I thought I might add a few thoughts, comments, or clarifications to his post.

  • Aaron mentions that Hyper-V is paravirtualized. Well, sort of. Hyper-V does support a paravirtualization interface (Hypercall, I believe?) for guest operating systems (i.e., Linux) that support it. In addition, future Windows guests will be “enlightened” as well. The confusing part about this is the fact that Microsoft (and Citrix, too) use the term “paravirtualization” to refer to the use of paravirtualized drivers instead of referring to the guest OS itself. Paravirtualized drivers are really nothing more than virtualization-optimized drivers, and it’s possible to use paravirtualized drivers even when the guest OS has no idea it’s being virtualized. In my mind, that’s not the same as “true” paravirtualization. Note that VMware ESX supports VMI, another paravirtualization interface, for guests (i.e., Ubuntu Linux) that support it. Also keep in mind that every major virtualization vendor offers optimized/paravirtualized drivers, including VMware, Microsoft, Citrix, Virtual Iron, and Novell.
  • You’ll also see the base Windows Server 2008 (or Server Core) installation referred to as the “Parent Partition.” All I/O travels through this installation. Whereas ESX uses a direct I/O model (drivers embedded in the virtualization engine), Hyper-V uses indirect I/O (drivers residing in the parent partition). Each side thinks their approach is the best, of course.
  • Aaron makes some comparisons between Quick Migration and VMotion, which is understandable but not entirely appropriate. Quick Migration is not live migration, but is really more comparable to VMware HA. Quick Migration has some advantages over VMware HA (can be configured on a per-VM basis), but also has some disadvantages (requires a dedicated LUN for each VM for which Quick Migration is enabled). I’ve discussed Quick Migration vs. Live Migration before.
  • Aaron briefly mentions SCVMM and expresses some doubt regarding using SCVMM 2008 (currently in beta, due to be released Q4) to manage VI3. It’s certainly possible, and it does require VirtualCenter in order to work. Whether it’s the home run that Microsoft hopes it will be is another story.

Thanks to Aaron for providing a good overview of Hyper-V. For more information, I’ll refer readers to some of my Hyper-V and SCVMM session liveblogs from Tech-Ed back in June:

VIR367: Hyper-V Security and Best Practices
A Discussion with Jeff Woolsey
VIR253: Microsoft System Center VMM 2008, Part 1 of 2
VIR360: Microsoft System Center VMM 2008, Part 2 of 2
VIR250: Advanced Storage Connectivity for VMs
VIR358: Hyper-V Architecture, Scenarios, and Networking
VIR350: System Center VMM Advanced Integration

For even more resources, readers can also use the HyperV tag to browse the site for all articles tagged for Hyper-V.

Oh, and if you don’t have Aaron’s RSS feed in your RSS reader, you’re missing out. He’s producing some really great stuff that you need to be reading. Go subscribe now!

Tags: , , , , ,

« Older entries § Newer entries »