When configuring your vDisks in Citrix Provisioning Server, you’re able to define which mode you would like your vDisk to be in. While many people know the basic differences, not everyone understands how to leverage them in their environment.
Private Mode
Private mode puts the vDisk into a volatile state where a target device can make permanent changes to the vDisk. Changes are not stored in a write cache file, they are changed immediately. When a target device connects to the vDisk, it gains exclusivity and no other device may connect to that vDisk.
This mode is best used when a single user needs application install capabilities which need to persist through reboots and updates. This is idea for streaming an operating system to a physical target device in order to move their hard disk to the datacenter for data consolidation. This mode is also used to make changes to a vDisk that is otherwise in Standard or Difference mode.
Standard Mode
Standard Mode is the bread and butter of Provisioning Server. It allows for changes made by the user, but they do no persist through reboots. They are stored in a write cache file and cleared upon graceful shutdown. The location of the write cache file is defined on a per vDisk basis and can be placed on the Provisioning Server Disks, in the target device’s RAM or local disk – with options for encryption and maximum size (when placed in RAM).
Standard Mode is usually used when leveraging XenDesktop so that users receive a standardized desktop which they can make temporary changes to without affecting the experience of other users. This can also be used when streaming an OS to a desktop to turn it into a pseudo thin client.
Difference Mode
Difference Mode is an interesting mode. It’s not often used, as far as I know. It’s essentially a hybrid standard-private vDisk. The base of the image is standard, but the user can make changes which are stored in a write cache file just like standard, but they persist through reboots. When the target device is booted, it boots to the standard image and then the write cache file is applied like a delta file. The main issue with this is when a change is made to the standard base, it destroys the delta write cache files.
Possible uses include tech labs and classes where a standard environment must exist for all users, but users need to make changes as part of the curriculum. The instructor can then remove the write cache files and “reset” the environment. A serious time saver in that case.
Conclusion
I only listed a few short use cases in this post. I’m really curious how everyone leverages these modes in their environment – please share!
Citrix Provisioning Services (PVS) is a great tool that is what our local Citrix Sales Engineer calls “the Secret Sauce” behind XenDesktop. For those who are new around these parts, PVS is software that allows us to stream an Operating System to a PC, physical or virtual, ESX or Hyper-V or XenServer – anything that can PXE boot pretty much (including Thin Clients!).
Once you’ve made the plunge and start deploying or planning your PVS infrastructure, High Availability and Redundancy is going to be a top concern. Here are a few guidelines to help you plan. Let’s assume we have a single farm, single site, with 2 Provisioning Servers.
Shared storage is your friend with Provisioning Server HA.
vDisk Storage
Ideally, this should be on shared storage. Server should be able to communicate with the SAN/NAS/etc at high speed. Fibre Channel works great here (4Gbps). Alternatively, you can have identical vDisks available via the same drive letter on each of the servers.
Write Cache Location & Storage
The two types of write cache locations that are in play: Stored on PVS Disk and Stored on Client HD (ignore Encryption, it’s not really relevant here). Storing the Write Cache on the Client’s Hard Drive is fastest, and is what I would recommend, as the file is available independent of the provisioning server. Alternatively, you can store the write cache on shared storage that can be accessed by each server so that the target device can continue to function.
TFTP Server
This is the tricky part. The best way to handle this is with hardware load balancers, but for the sake of this article, I’ll give you some “rigging” solutions. You can run the TFTP service on both servers and use “round-robin” settings on your DHCP server. Also, you can use the Bootable Image tool included with PVS to create a Bootable CD that contains the bootstrap file (the same one that it can get via PXE). If you have a NetScaler in your environment, you can use it to ensure HA of TFTP (Thanks Bart Groot Zevert). I’m sure there are other ways as well, feel free to comment with suggestions.
Conclusion
As mentioned before, these are just some guidelines and tips to get you started on the right path. I’ll be more than happy to assist should you need assistance.
Planning for your new XenDesktop environment can be a tough process. Being a relatively fresh technology, there aren’t too many places to find straight answers regarding scalability. I intend on changing that. Here are a few helpful tips and straight answers to help you decide whether you have the hardware already, or need to purchase new. This will be a multi-part series, only some aspects will be addressed in this initial post.
Assumptions
While this scenario is pretty simple, I’ll be more than happy to help you with your questions in the comments, via Twitter, or e-mail.
In the next post, I’ll discuss Virtual Machine Specifications and choosing the right server specifications for your XenServer hosts.
In the past year we’ve seen an influx of new product offerings from Citrix, VMWare, Microsoft, and other smaller vendors. We’ve also seen the “Cloud” take off. Many people say these new technologies are simply hype, or are useful in only a handful of environments. I think they’re wrong about both ideas.
New technologies such as the Client Hypervisor, the cloud, and even VDI are all wonderful new things; begging to be taken advantage of. Things change so fast that we can’t think in the same way too long, we have to have malleable minds. Don’t think about where you would utilize a client hypervisor in your current environment – you’ve already, most likely, perfected it without this new technology. You have to think about how you can make your current environment better with this new tool. You have to re-evaluate every decision you’ve made concerning your current environment.
In order to make imaging our PCs more efficient, we are replacing all of our PCs every 3 years in order to make sure they are not only up to date hardware-wise, but so that they are all the same and share drivers.
This scenario is more than likely all too familiar to many of you. With the client hypervisor, this can change, perhaps for the better.
We can increase our hardware lifecycle by using Citrix Provisioning Server to stream virtual disks to the PCs. With a client hypervisor, we dont have to have a standard model PC, we can replace computers as necessary and at a pace that is more cost efficient.
Both of these scenarios work. Which one do you prefer?
Microsoft recently announced its inclusion of a “Windows XP Mode” for Windows 7. Users will be able to run their applications in a virtual machine running Windows XP SP3. For us virtualization folk, this is pretty neat that virtualization will make it to the average user, but is it the wrong way to handle compatibility issues?
Short Answer: Yes.
The best way to handle the compatibility issue, according to many journalists, is to use Microsoft App-V to virtualize just the application and the files it requires, not the entire OS. Managing a second file system, registry and driver set will be tedious and more than likely, won’t work very well.
My idea? Don’t bother with it at all. Software Developers have lagged behind since Vista was announced in 2005. Vista flopped in part because of the developers failure to participate. Do we really want to allow these people to program assuming everyone in the world can still run XP? When will we ever move forward?
I was chosen as one of the seven engineers to assist in the development of the new Citrix Certified Enterprise Engineer™ (CCEE) certification exam. After the week-long workshop, I’ll be accredited as a Subject Matter Expert in Citrix Virtualization Technologies. The workshop is April 20th to the 24th in Ft. Lauderdale, FL.
Participants in the workshop will:
- Receive training on how to write questions for certification exams
- Work alongside our Exam Developers to develop questions
- Work alongside other field experts to review, modify and finalize questions written by other participants of the workshop
At VMWorld Europe, Citrix announced that their popular virtualization platform, XenServer, would be offered free. Virtualization is quickly becoming standard practice at the enterprise level, and for those who don’t currently virtualize, being able to do so for free really helps the process.
XenServer is different Desktop Virtualization software, such as Sun’s VirtualBox or VMWare’s Workstation, in that it is a true hypervisor. It runs a stripped down version of CentOS Linux which emulates hardware at a very low level. Desktop Virtualization software usually requires a full OS, such as Windows XP or Ubuntu Linux, in order to run.
You might expect a vendor to cripple their free offering, but Citrix has done well – they left plenty of the great features intact. For example; XenMotion is a feature that allows you to move a VM from server to server without interruption of service (NOTE: Requires shared storage – e.g. SaN). A more detailed list of features can be seen here.
To use XenServer, you’ll need dedicated hardware. Simply download and burn the ISO, boot to it and follow the prompts. Then, install XenCenter – the software that you’ll use to manage the pool, on your workstation.
More information can be had at Citrix.
Develop a Plan to Migrate from Windows to Open Source Software
It’s unrealistic to think you will shed your Microsoft addiction overnight, or even over a year, but developing a plan to move a good portion of your infrastructure to Open Source, will save a noticeable amount. The caveat is finding staff familiar with Windows to UNIX migration and administration.
Be Energy Conscious
Consider power usage and efficiency when weighing hardware purchases. Look for high efficiency power supplies when deciding on workstations. While environmentally friendly hardware may be a bit more expensive initially, it will pay off over time. Ensure your datacenter is properly insulated so that the conditioned air remains cool and where it belongs. Consult a specialist in Green IT solutions if needed, they can do the math and quantify your expected savings. Try Google’s PowerMeter software.
Server Virtualization
Virtualization may be the most effective way to save money. Bare-metal servers rarely sustain >%50 CPU usage, leaving the other half unused. Using a Type 1 hypervisor, you can split the hardware up into smaller pieces so that the server can be utilized more efficiently. I recommend Citrix XenServer or VMWare’s ESX.
Knowledge Bases
Creating a central knowledge base with easy to access and easy to understand instructions to help users resolve common issues will help reduce dependency on your support staff, allowing you to allocate them elsewhere.
Desktop Virtualization
The average life span of a desktop is 3 to 5 years. Beyond that and things start breaking – and everyone in IT knows you should replace before it breaks, not when. Creating a Virtual Desktop Infrastructure (VDI) will offer your users a real desktop, a virtual machine on a server in your datacenter. A thin client is used to access it. Thin Clients’ life span is 5-7 years depending on the model and are significantly less expensive than their big brothers.