VM /ETC

Posts Tagged ‘design’

By now you’ve seen the pictures, video (VMworldTV), and posts about the hardware in the datacenters that hosted the VMworld 2009 Labs. You should already know about the staggering number of virtual machines ( > 37,000 ) running on the ESX 4 servers ( > 770 ). But enough about the hardware.

If you are like me you probably would have loved to get the opportunity to use the vSphere client to connect to a vCenter server managing that entire virtual infrastructure (VI). Although I did not get to do just that, I did get the opportunity to do the next best thing – talk to the manager of the team that does. My VMworld ended by talking to Randy Keener, Group Manager of VMware’s GETO team (Global Engineering Technical Operations). Keener explained to me some of the VMworld 2009 virtual infrastructure design details that VI administrators would be interested to know.

Nested ESX in the Lab Manager Cloud

What Keener revealed somewhat surprised me. Although vCenter 4 server was a piece of the design, the true magic that supported the self paced labs, instructor led labs, and the Solutions Exchange was (arguably) an example of a private cloud created by

Read the rest of this entry »

Jason Boche’s post titled Idle Memory Tax is a great read if you are trying to understand ESX memory allocation between virtual machines (VMs). Specifically, the post does a great job explaining how it works when you over commit your physical host’s memory. In other words, the sum of all the RAM assigned to the VMs running on a host is greater than the actual physical RAM of the ESX server.

Here’s a quote from Jason that briefly explains part of the technology that makes over commit possible: the Idle Memory Tax (IMT).

“Quite simply it’s a mechanism to take idle/unused memory from guest VMs that are hogging it in order to give that memory to another VM where it’s more badly needed. Sort of like Robin Hood for VI. By default this is performed using VMware’s balloon driver which is the more optimal of the two available methods. Out of the box, the amount of idle memory that will be reclaimed is 75% as configured by Mem.IdleTax under advanced host configuration. The VMKernel polls for idle memory in guest VMs every 60 seconds.”

Read the entire post for much more technical details and examples.

I’ve blogged before about the symptoms when the IMT and the ESX balloon driver can no longer keep up and it’s time to add another ESX host and spread the VM load.

I believe that ESX 3.x changed the need to Read the rest of this entry »

Back in December Scott Herold posted on the VMGuru.com web site that free chapters from the popular book “VMware Infrastructure 3: Advanced Technical Design Guide” would be available for download every 2 weeks. Eventually the entire book would also be available as a down loadable .pdf. That’s great news, but as of this writing I could not find any of the chapters on the page of downloads available from the VMGuru.com site.

Scott, Mike, Ron, consider this post a reminder that it’s mid January now (I know you guys have nothing else going on ), and hopefully this gives you some added promotion and excitement about your book. I’m sure a lot of VMware Admins will want personal copies.

Read Scott’s original post  titled Free Chapters Available Starting in January.

“Ron, Mike, and I have decided that starting in 2009, we will be giving out 2 free chapters of the VMware Infrastructure 3: Advanced Technical Design Guide every 2 weeks.  I am still in the process of collecting the source DOC and PDF files from the Publisher, but as of right now we are anticipating the first two chapters to be availabe during the first week of January. “

There are 0 comments on Scott’s post so maybe he thinks no one was listening? Help me show him we are all waiting!

If you can’t wait for the downloads then Read the rest of this entry »

iSCSI – Hardware or Software – How many TOEs do you have? is a post by Carlo Costanzo that really struck a chord with me. Carlo asks:

“More and more of my new implementations of VMware Infrastructure are being connected to iSCSI SANs (EMC, LeftHand, and Equallogic) and the question has come up about whether or not to spend extra dollars on TOE (TCPIP Offload Engine) Network cards.”

This made me realize that I have yet to implement an iSCSI SAN connection to an ESX Cluster with a hardware initiator. My customers have all used the native ESX iSCSI software initiator instead. So it made me wonder why am I not even considering the hardware iniator anymore, and should I present the option to my customers at all? Afterall, as Carlo points out, the number of supported TOE cards has increased from 2 to 16 (a table is provided at the end of this post)

I came up with a short list of reasons why I haven’t been using the iSCSI hardware initiators. Read the rest of this entry »

I was surprised to discover that the ability to estimate the size needed for the VMware VirtualCenter (VC) database was built in to the VirtualCenter Management Server Configuration options in VC 2.5. This means that the next time you are asked by your DBA “How big a database does VC need?” you can answer that question from the convenience of your VI Client.

Just point the VI Client at your VC 2.5 server and then use the Administration MENU (not the administration button). Select VirtualCenter Management Server Configuration. Then click on Statistics.

Here’s a screen shot from a VI Client pointed at a VC 2.5 Update 3 server. Read the rest of this entry »

Look for a new VMware KB article in the near future for this VMware VDM error.

I helped implement a VDI solution for a customer that ended up revealing an issue with VMware Desktop Manager (VDM) 2.1 and the resource pools of an ESX Cluster. The explanation of the problem to me was that the VDI desktops auto-deployed via persistent VDM pools were out of sync with the ESX resource pools they were members of, and therefore new VDI desktops could not be cloned by VDM. That’s a little confusing I know, so I’ll try to explain it better by providing the sequence of implementation steps that resulted in the issue.

Here’s how it happened. Read the rest of this entry »

For my project, 5 TB of raw disk space on a NetApp FAS3020 resulted in roughly 2.5 TB of usable NFS volumes. Sacrificing half the storage as overhead (for both ESX and NetApp) might seem like a big hit to take at first, but when you consider what you are able to do in the remaining useable space it can change your opinion. In my recent experience, after combining the features of Flexclones, VirtualCenter (VC) templates, deduplication, and NFS the remaining 2.5 TB enabled a virtual desktop infrastructure (VDI) storage solution for over 80 virtual machines (VMs). Your results may vary, but the concepts discussed in this post provide for a scalable and flexible design.

Although this post discusses NetApp, I imagine any NFS filer with similar features could be used in this same fashion. I’m curious for feedback on not only this solution, but the pros and cons of other vendor’s features in a similar scenario.

Here’s a logical diagram illustrating a NFS volume strategy followed by some high level details of each volume. Read the rest of this entry »