Kenneth is Speaking at BriForum Denver

I sat down with one of Lewan’s Solution Architects, Kenneth Fingerlos, to discuss his upcoming speaking engagement at the BriForum conference on July 20th.  Our brief conversation covered the details of his session, “vSGA, vDGA, vGPU, and Software – When and Why“, his background in the industry, and what gets him excited in the technology space right now.

Me: Kenneth, can you tell me a little bit about your industry experience?

Kenneth: So, after college I took a left turn in my career path and went into corporate IT for ten years.  Various positions: desktop management, server management, data center.  Various kinds of things.  After ten years of that I decided I didn’t care for IT management and tried to correct the course change and landed in consulting.  I’ve been doing IT consulting for about the last ten years around storage, data management, virtualization of various types, and building up my skill sets trying to help customers solve problems.

Me: Great, great.  So have you been to BriForum before?

Kenneth: I have not been to BriForum.  This will be my first year.

Me: What attracted you to BriForum?

Kenneth: I’m excited.  The whole idea of a conference that has some size to it and is established that is not tied to a specific vendor is just exciting, right?  You go to a Cisco conference and it is all about what is the latest widget from Cisco.  Cisco can do no wrong.  You find the same thing if you go to, you know, Dell World.  Dell is perfect.  Whatever Dell has got going is awesome and whatever everyone else has is garbage.  BriForum excites me because it is everybody. It is a marketing company–a media company that puts on the conference as opposed to a product manufacturer.

Me: So what will you be discussing at BriForum?

Kenneth: I’m discussing a topic that is near and dear to my heart which is the idea of virtualized graphics.  Taking things we do everyday in the physical world with physical PCs and trying to bring this into this virtualized environment.  Things like disaster recovery, security, flexibility.  You know, the physical world is pretty restricted.  Graphics have always been one of these things that is hard and is difficult.  Technology is evolving and has advanced dramatically over the last couple of years in terms of what we can do.  But there is also a lot of complexity and a lot of information and I find my customers have a lot of confusion about what they can and can’t do.  What works, what doesn’t work.  My session is all about trying to bring some clarity to that area.

Me: Ok, so I am going to open this up a little bit and say maybe don’t limit this to just the enterprise world but what is the technology you are most excited about right now?

Kenneth: The technology I am most excited about right now….I think the stuff that is most exciting is really this idea of graphics virtualization.  I mean, so many things go into a user experience, right?  And all of the traditional things that you think about: servers, storage, memory, CPUs–graphics is part of that.  Remoting protocols, right?  What’s going on with actually getting that content delivered to a user.  Networking, right? 3G, 4G networks and starting to think about what’s next, what’s beyond 4G.  These are huge enablers to let people consume and develop content in ways that have never been envisioned before.  Letting you take that stuff to the cloud, to the remote data center, and access it from anywhere.  I’ve been sitting on top of a mountain in my 4×4 holding a virtual desktop, just because I’m a geek and into this stuff, but yes–I can access that app, whatever it is, from a mountain top in the middle of nowhere.  That’s cool stuff. And it’s all about enabling people to work and function in ways they’ve never been able to before.  That excites me.

Me: Very cool.  Well, looking forward to seeing your session at BriForum!  Until next time.

As I wrote about earlier, BriForum Comes to Denver, and I am excited to have such a great event in my backyard.  If you are going to be at BriForum or just have general questions about Denver, reach out to either @kfingerlos or myself (@sagelikebrian) and let’s catch up.

Brian @sagelikebrian

BriForum Comes to Denver

IT conferences are a great way to catch up on what is new, take classes, and network with peers in the industry. I have been lucky enough to attend great shows like Citrix Summit and Synergy as well as VMware VMworld over the years. The conference for me that always fell just out of reach was BriForum. This year it is all going to change. I am more than a little excited that one of the world’s premier IT conferences has chosen Denver, Colorado for this year’s US location. BriForum is an independent conference that provides vendor-neutral perspective on current and emerging technologies and services.

Check out this year’s list of sessions:
http://www.brianmadden.com/blogs/gabeknuth/archive/2015/03/09/check-out-the-list-of-sessions-for-briforum-denver-2015-july-20-22.aspx

If you have a keen eye, you may have noticed a third of the way down the list a special session, “vSGA, vDGA, vGPU, and Software – When and Why“, being presented by Lewan’s very own expert speaker Kenneth Fingerlos (@kfingerlos).

Kenneth will be talking about the new graphics intensive workloads that are possible in VDI thanks to highend GPUs from NVIDIA.  He will specifically be digging into the different methods you can use to virtualize the GPU and when and why you would want to choose each method.  I promise you this will be a deep technical dive preparing you for your next graphics intensive virtual desktop project.

Check out the Lewan IT Solutions Technical Blog for more great technical information from Kenneth.

Come join Lewan at BriForum 2015 if you would like to learn more about solutions from Citrix, VMware, Microsoft and much more.

Brian Olsen (@sagelikebrian)

Why is a smaller number of virtual CPUs better?

Note: This article is designed to serve as a high level introduction to the topic and as such uses a very basic explanation. Papers for those that wish to dive into more technical details of the topic are available elsewhere.

In a virtual environment such as VMware or Hyper-V, multiple virtual machines (VMs) operate on the same physical hardware. In order to make this function, a small piece of software, called a hypervisor operates to schedule the virtual resources with the physical hardware. As a virtual machine enters a state where CPU resources are required the VM is placed into a CPU ready state until enough physical CPUs are available to match the number of virtual CPUs.

The hypervisor will schedule VMs to available physical resources until all resources that can be scheduled are used.

Each VM will run on the physical CPUs until either it needs to wait for an I/O operation or the VM uses up its time slice. At that point the VM will either be placed into the I/O wait state until the I/O completes or be placed back in the ready queue, waiting for available physical resources.

As physical resources become available, they hypervisor will schedule VMs to run on those resources. In some cases, not all physical resources will be in use, due to the number of virtual CPUs required by the VMs in the ready state.

The process continues as VMs either wait for I/O or use their time slice on the physical CPUs.

In some cases there are no VMs in the ready state, at which point the scheduled VM will not time out until another VM requires the resources

Often a VM with fewer virtual CPUs will be able to be scheduled before one with more virtual CPUs due to resource availability.

In some cases a VM will complete an I/O operation and immediately be scheduled on available physical resources.

Algorithms are in place to ensure that no VM completely starves for CPU resources but the VMs with more virtual CPUs will be scheduled less frequently and will also impact the amount of time the smaller VMs can utilize the physical resources.

A VM with high CPU utilization and little I/O will move between the ready queue and running on the CPUs more frequently. In this case, the operating system will report high CPU utilization, even though the VM may not be running for a majority of the real time involved.

In these situations, operating system tools that run within the VM may indicate that more CPUs are required when, in reality, the opposite is actually the case. A combination of metrics at the hypervisor and at the operating system level is usually required to truly understand the underlying issues.

LoginVSI “VSISetup has stopped working”

One of the real challenges of testing a Virtual Desktop Infrastructure (VDI) is getting enough users to logon to the system to test it at the same time.  While load testing should be an essential part of validating a new system, scheduling three or four users to test applications can be challenging.   What if you would like to simulate the load of 25 or 50 users?  Often, due to the challenges of scheduling and man hours involved load testing gets ignored and you end up hoping for the best on go live day.

LoginVSI is an application that allows you to orchestrate as many users as you would like to test the system while having it all be automated.  Out of the box, LoginVSI will simulate a user logon and then perform a series of very real activities like surfing the web, reading emails, or editing spreadsheets.  This simulated workload progresses for about an hour per user.  This allows you, the administrator, to observe how resources are used and look for issues related to a stressed system.  LoginVSI can be set up to keep having new users logon until the system stops responding or crashes.  It then compiles the performance metrics and tells you the optimal amount of users the system can handle before the server starts to have poor response time.  They call this number the VSImax.

This article explains how to resolve the error, “VSISetup has stopped working” while setting up LoginVSI.  During the install process of the LoginVSI management console the setup program crashes shortly after starting it.

Resolution:
Add the .NET Framework 3.5.1 using the Add Features wizard.  Before continuing on with the LoginVSI install, run Windows Update and patch .NET.  This can be time consuming as there are lots of updates available and this may require a reboot or two.

Cause:
While it is documented in the excellent install guide, it is easy to forget that the default install of Windows Server 2008 R2 does not have .NET 3.5.1 installed and it is required.

Brian Olsen @sagelikebrian

Unable to get vCenter server certification chain error during vcOPs 5.8.1 install

During the deployment of the vcOPs vApp for a customer I ran into a new error – well, new for me. While the vApp (v5.8.1) deployed and booted fine, as I was registering it with the vCenter (v 5.1) as part of the initial configuration I got the following error: Unable to get vCenter server certification chain. Off we go to Google… Here’s a quick summary of things to check:

• Confirm name resolution is working, username/passwords are right, etc.
  • Assuming Windows, RDP to vCenter with the user you’re attempting to use or try access via your favorite vSphere management tool
  • Hop on the console of the UI VM and ping the vCenter by IP and DNS name (username: root initial password: vmware)
• Check that your vCenter certificate hasn’t expired. It’s the rui.crt file in c:\ProgramData\VMware\VMware VirtualCenter\SSL. This article has good info on locating and renewing your certificate, should that be your problem.
• In the end, my fix came by importing the certificate file to the UI VM manually as outlined in this VMware KB article.
  • Full disclosure, the symptoms in the above article didn’t match my problem exactly and I don’t like just trying random fixes. However, when I found this Blog Post, in Spanish, with my exact error recommending a similar .cert import process I threw caution to the wind. The exact steps from the Spanish blog didn’t quite work, which could be a result of my inability to read Spanish and/or Google Translate not being perfect, but the VMware KB article was spot on.

After importing the certificate manually and restarting services, all was well and I was able to complete the configuration of vcOPs. By the way, did you know that since vSphere 5.1, all licensed versions of vSphere now include the Foundation edition of vcOPs? More than 5 hosts in your environment and you’ve got enough scale to warrant leveraging this tool. For a limited time, VMware is letting Lewan perform a free vSphere Optimization Check including a 60 day trial of the Standard Edition, complete with the capacity management features, dynamic thresholds, and root cause analysis. Give us a call today to test drive Operations Management!

Thinking about a VDI initiative? Watch this.

Lewan Solutions Architect, Kenneth Fingerlos, wowed the crowd last month at the GPU Technology Conference (GTC) 2014 with his presentation on VDI, “Virtual is Better than Physical Delivering a Delightful User Experience from a Virtual Desktop“.

GTC is the world’s biggest and most important GPU developer conference. Taking place in Silicon Valley, GTC offers unmatched opportunities to learn how to harness the latest GPU technology, along with face-to-face interaction with industry luminaries and NVIDIA experts.

Leveraging his industry leading expertise, Kenneth “delivered in spades,” as described in a review of his presentation for The Register:

The VDI talk was the kind of GTC session I love. It’s where a real-world expert talks about how a difficult task is actually accomplished. Not the theory, not how it should work on paper, but what it takes to actually move a project from Point “A” to Point “We’re done with this”.

Ken Fingerlos from Lewan Technology delivered in spades with his “Virtual is Better than Physical: Delivering a Delightful User Experience from a Virtual Desktop” GTC14 session. Delightful? Hmm…In my past lives, I’ve had to use some virtual PCs and my experience ranged from “absolutely unusable” to “omg I hate this”.

It’s easy to see that Fingerlos has been around the block when it comes to VDI. He has all the right credentials, ranging from VMware to Citrix to Microsoft. But more importantly, he’s been there and done it.

Read the complete review from theregister.co.uk

Kenneth’s GTC Presenter’s Bio

View the complete session and slide deck:

 

Windows Server 2012 Licensing – a quick reminder

This came up recently for a customer and while it’s not new news, I thought a quick reminder would be useful. There are a few key points to remember about licensing of Windows Server 2012 in server virtualization projects, these rules apply to XenServer, VMware, Hyper-V, Oracle VM, etc.:

• Licenses are applied to physical servers, never to virtual machines. If you are thinking about how you need a license for the VM you are about to build, you’re probably doing something wrong
• There is feature parity between Standard and Datacenter editions, Enterprise Ed has been dropped
  • The only difference between these 2 major editions is in the number of virtual OSE’s (operating system environments, aka a virtual machine) granted with the license
  • A license covers 2 processor sockets within 1 server, 1 license cannot be purchased to cover 2 servers each containing 1 populated processor
  • The license allows for one bare-metal install of the operating system, but doesn’t require it – as would be the case if your hypervisor is anything other than Hyper-V
  • Virtual OSE grants by edition:
    • Standard: 2 virtual OSE’s per license
    • Datacenter: unlimited OSE’s per license
• More than 1 license of the same edition may be applied to a given physical server to cover additional CPU sockets or additional virtual machines
  • 2 Standard Edition licenses would cover 4 processor sockets and/or up to 4 VM’s
  • 2 Datacenter Edition licenses would cover 4 processor sockets and two * unlimited for the number of VM’s ..that’s like beyond infinity, but 4 CPU sockets.
• The license cannot be transferred more than once every 90 days – yeah, you read that right. This rule is to prevent a license from jumping from one host to another to follow live migration activities
  • This is where most people pause and say “oh..”. That tells me they were purchasing 1 license per VM and just thinking the license moves around with the VM
  • You need to cover the high water mark of virtual OSE’s for a given host
• Licensing math:
  • Standard Ed. list pricing is $882
  • Datacenter Ed. list pricing is $4809
  • The break-even point for Datacenter is at 5.45 Standard licenses; in effect, for a density of more than 10 VM’s (5 std licenses each granting 2 OSE’s), you should use a Datacenter Edition license
• A real world example: New virtualization customer deploying 3 VMware hosts
  • We generally size the environment for N+1, meaning we’re planning that 1 of the servers is a “spare” from the perspective of workload sizing – so all the workload can run on just 2 servers; we’re planning for this and so should you in your licensing.
  • If you plan to run more than 20 total VM’s in this environment, you need 3 Datacenter Edition licenses
    • 20 VM’s running on 2 servers = 10 VM’s/server
    • 10 VM’s requires 5 Standard Edition licenses to have enough OSE grants
    • More than 10 per server, and it’s now cheaper to have just bought a single Datacenter Edition license
      • 6 * $882 = $5292, which is greater than $4809 for datacenter
    • Since you don’t know which host (think of a rolling patching cycle) is going to carry the increase load, all the hosts in the environment should be licensed uniformly to this high water mark
  • Depending on the licensing model, an upgrade from 5 * Standard Edition licenses to a single Datacenter Edition license may not be possible – plan ahead!
  • If you have OEM licenses that came with your old physical server environment, these are likely not transferrable – they don’t follow the P2V action
• With this understanding, while you might have some work to do upfront (or scrambling to get back into compliance now) the long term savings are very real for dense virtualization projects that can leverage the Datacenter Edition license. On a modern 2 socket server with 16 cores/32 threads, 10 VM or greater density is easily achievable

General licensing FAQ:
http://download.microsoft.com/download/4/D/B/4DB352D1-C610-466A-9AAF-EEF4F4CFFF27/WS2012_Licensing-Pricing_FAQ.pdf

Licensing brief for virtualized environments:

http://download.microsoft.com/download/3/D/4/3D42BDC2-6725-4B29-B75A-A5B04179958B/WindowsServer2012VirtualTech_VLBrief.pdf

Exchange Infostore will not start because of Vmhost time setting?

We recently had several clients whose exchange servers went offline for seemingly no apparent reason.  The servers were up and there were no connection or permissions issues, but we kept getting a strange error in the logs.

Error: Unable to initialize the Information Store service because  the clocks on the client and server are skewed.   This may be caused by a time change either in the client or the server,  and may require a reboot of that computer.   Verify that your domain is properly configured and  is currently online.

After investigation, we found that the Vmhost running the exchange server did not have the correct time set up.  We changed the time on the host (we actually enabled NTP to match the domain controller), and rebooted the exchange server.  The exchange server came up and the infostore started without error.  Everything was then fine in the world again.

Related articles

• PowerShell: Identify the number of users connected to a Exchange 2010 CAS Server (oxfordsbsguy.com)
• Check you Vmware load (b3itonms.wordpress.com)

Lewan Achieves Veeam Gold/Platinum Partner Status

Our Enterprise Solutions team has been hard at work in the lab, training to become solutions experts for Veeam Data Protection and Backup tools. Their hard work, dedication and opportunity to train with Veeam’s technical team has earned Lewan the recognition of Gold Partner status in the Veeam ProPartner Program. Congratulations!

View all of Lewan’s Vendor Partners: http://www.lewan.com/vendorpartners

Proud to have CU as one of our customers

University of Colorado Boulder Housing and Dining Services Goes Green With New Virtual Desktop Infrastructure

Hybrid Storage from Nimble and Scalable Cisco UCS Servers Deliver Solid Foundation for VMware View VDI

Click to access nimblestorage-cs-cu-boulder.pdf