70 million individual dollars can buy you a lot of things. A 64 metre long super yacht. The services of an NFL linesman for two years. For $70 million you could entice an English Premier League striker to play for you, but not necessarily score goals. $70 million is 113,000 Apple iPads. If you spent $100 a day, it’d take you nearly 1,950 years to get fritter it away. Yet despite all these glittering prizes and goals, Goldman Sachs chose to invest their $70million in a chunk of AppSense.
Of all the things they could have invested in, why did choose AppSense? If the future is going to be full of cloud services, virtualised desktops, and mobile devices, why spend a not inconsiderable sum on something that sounds the stuff of science fiction?
What is User Virtualization and is it worth a $70 million dollar investment? Why would you need user virtualization? And indeed what makes AppSense stand out?
What is User Virtualization?
User Virtualization makes your user’s information manageable and portable. With User Virtualization the components of a desktop relating to the user are decoupled from the operating system and applications. This allows them to be managed independently and applied to a workspace as needed without scripting or group policies, regardless of how the workspace is being delivered.
Why is User Virtualization useful?
To best understand what User Virtualization enhances, it’s useful to define what the default option is. If your users have a workspace within any windows desktop environment (be that PV, VDI, blade PCs or traditional desktops) each user will have their own settings that define how their applications work and what resources they access. They’ll work within a security framework that grants (or denies) them rights to change these settings and resources. Microsoft’s term for managing this set of settings is “User State Virtualisation” This is in fact, a fancy, re-branding and re-badging of three different technologies:
1.Roaming Profiles: Every windows user has a set of characteristics for the way they use and view applications and their workspace. This is their profile. It is essentially a database of application settings for the user and a set of files. You can have a “local profile” i.e. your settings are confined to one device: log on to another device and your application settings aren’t available and you have to set them up again. You can have a ‘roaming profile’, your settings are saved and loaded from a network location when you log off/on to a device. This allows you to ‘roam’ between PCs – with your settings being installed as part of the log on process. You can also have a mandatory roaming profile where essentially – your settings are defined by someone else and are not changeable. The use of mandatory profiles tends to be for specific-task workers.
2.Folder Redirection: Within a roaming profile, the number of files within that profile can become large over time. Transferring those files each time the user logs on, or off, can be time consuming. By redirecting folders to a central location you can avoid this save/load process – and reduce logon/off times. However, you also increase your reliance on a good network and file storage infrastructure: get this wrong and you can degrade the user experience as each file access takes longer.
3.Off-line Files: you can make shared network folders available when the user isn’t connected to the network by using off-line folders. Off-line folder technology relies on synchronizing files between the device and a central network point.
At a recent user group meeting I was reminded of the fact that in 2010 the Windows Team Blog stated “The original design of Roaming User Profiles, Offline Files and Folder Redirection doesn’t really account for this newer type of work style”. By “newer type of work style” the Windows Team mean “anything other than a dedicated Windows PC workspace configured for a single user working on an office local area network“.
If your desktop strategy was to provide each user with a dedicated PC, with applications installed for them on that PC – you’ve likely not come across the deficiencies of “user state virtualization”. You’ve also likely thought ‘this way I deliver desktops is costly, cumbersome and needs to change’.
You will have come across the deficiencies of “user state virtualization” if you’ve tried to implement virtual desktops – especially pooled virtual desktops; or you’ve implemented a Presentation Virtualization environment; or you’ve had to implement a hot-desking/hotelling environment for traditional desktops: even more so, if those traditional PC users roam between sites.
What can possibly go wrong with User State Virtualization?
Without user state virtualisation, the core issue is that each time a user logs on to a device – it is as if that device is being used for the first time. Reconfiguring settings each logon instance is time consuming and wasteful.
Even with user state virtualization there are still issues. A log-on should take seconds – with poorly implemented user state virtualization this can take minutes.. or tens of minutes. This is because the amount of data transferred from PC to network can grow over time if unchecked and unmanaged. Disk may be cheap – but this wastes time. It wastes time not only for the user logging in/out – but for other users who are impacted because the file server or network is inundated with profile file transfers. Issues with applications not working correctly.
Why is this only a problem now?
Its not. Its been a problem for a long time. However there has been a greater trend towards changing the desktop delivery model to improve flexibility and to reduce management costs. Typically desktops were served on a dedicated 1 user-1 desktop model. Such a model is being challenged as companies look to manage their desktops differently – to mix traditional desktops with laptops, with hosted desktops, with presentation virtualisation.
User virtualization can be used to enable this greater agility, to reduce the time taken resolving issues with user’s application settings and reliability.
What Alternatives to User State Virtualization Are Available?
There are a number of vendors offer solutions for implementing user virtualization.
Both Citrix and Quest have user virtualization facilities within their desktop/presentation virtualization offerings. VMware are still yet to introduce their user virtualization function, a facility that has been in development since VMware acquired RTO Profiles (which at the time wasn’t even a user virtualisation offering). The issue here is that these vendors offer a solution that is closely tied to you having their desktop virtualisation offerings.
Vendors such as Immidio with Flex Profiles and Liquidware Labs ProfileUnity offer products that can be used across a variety of devices and need not be deployed with a VDI/PV solution. Their solutions don’t require a database to save settings so can be easy to implement and have a low operational cost.
On the other hand, vendors like AppSense, RES Software, Scense and Tricerat have a more complex architecture. They require you deploy and manage agents on devices, they require databases for configuration information. However, these additional requirements enable their respective products to have a great granularity of control over how and when changes are made, provide reports on activity, allow delegation of control and readily allow versioning and auditing of how, when and which settings were applied.
There is a difficulty in creating a comparison of user virtualization products because the feature sets often do more than “be more effective user state virtualisation”. But, if you were to make a sizable investment, AppSense offer a product set that is designed for a wide range of business sizes especially large enterprises with more complex environments; it has an extensive user base and partnerships with desktop solution vendors such as Citrix. Like other vendors, AppSense is not tied to a specific workspace delivery method but its marketing focus has consistently been about managing the user environment: not a transition or migration, not as a part of a wider management model. AppSense’s focus is on offering a consistent user experience, regardless of device.
Ask not what you can do for virtualising users, but what User Virtualization can do for you
The goal of user virtualization is to provide consistent and seamless working environments across a range of application delivery mechanisms, making the working environment predictable and responsive, simplifying IT administration and reducing costs.
Microsoft are not under any incentive to resolve the issues of user state virtualization. That core model fits a 1-1 user-device environment; and that 1-1 model drives PC/laptop sales, which in turn drives Microsoft OS license revenue. However, businesses have discovered that the 1-1 model is inefficient, costly and difficult to change. It relies on having a personal computer in a local office, with local servers. For many organisations, this requires big offices, and big bandwidth.
Successful user virtualization is key to changing a desktop delivery model: otherwise the only option is to maintain the status quo: do as you’ve done before with all that cost and inefficiency.
Is User Virtualization worth a $70 million dollar investment?
AppSense have done well in their original markets in Europe and the UK most specifically: wider desktop sharing and a higher presentation virtualisation market offered good opportunities. AppSense’s new launch into the key US market has been successful: it rode neatly on the fanfare of “!!hey everyone, look at VDI!!” because VDI implementations forced IT administrators to understand how limited Microsoft’s default offering was. While enthusiasm for VDI may have waned, AppSense’s product set offers savings and better management beyond VDI. Appsense’s combination of settings management, workspace configuration, application lock-down, user rights management and license control is applicable to a wide range of desktop delivery strategies. Goldman Sachs investment is not to gain market share per say, but to allow AppSense to rapidly expand to meet the demand that their product has generated.
Its always wise to invest in something you are sure is going to be successful.
Does that mean that AppSense will be the outright winner? No. No because other vendors offer viable solutions and $70 million isn’t going magically change the feature set making the product all things to all men. Indeed, Klaus Besier chairman and CEO of RES Software applauded the investment as ultimately it helps drive a rapidly expanding desktop virtualization and management market.