Why are organisations who want to Virtualise their desktops choosing VDI?

Image

As Business Development Director of Molten Technologies, I spend a lot of time with clients discussing what they want to achieve from Desktop Virtualisation and which of the multitude of solutions out there they should choose.

One Size never fits all

A consistent theme during these conversations is that a single solution will almost NEVER fit all the use cases in a single client. Even in relatively small deployments, there will be a number of different groups of people that use their desktops in a variety of different ways – accessing different applications and data, working from different locations and doing a wide variety of different roles. 

There are some very clever and useful technologies out there today, that can be generically grouped under the heading ‘Desktop Virtualisation’. Solutions such as application virtualisation, application streaming, desktop presentation, local hypervisors and (coming soon) rich web applications such as HTML5, can be excellent solutions for the right use cases. 

It is my supposition however that, today – now, VDI is the right solution for most of the people, most of the time.

So in the real world, why are organisations choosing VDI?

Given that an organisation probably still has legacy applications that require it to provide (usually Windows) desktops, why choose VDI over the alternatives?

Clearly different organisations have different priorities. It is however possible to identify some consistent themes from those organisations who have taken the plunge and built, or bought, VDI.

1. ‘Normal OS’ with ‘Normal apps’ – but in the DC, without changing anything else

It comes down to the real world of IT management in most organisations – they typically have A LOT of applications, some of which are very specialist and often quite old. In an ideal world these applications would have all been migrated to newer platforms years ago – but in the real world people have built processes and tools to manage them. Whilst not perfect, these management systems work and are a known problem.

VDI divides the problem of change into bite sized chunks, offering an organisation the opportunity to get the great benefits of virtualisation (remote access, security, cost … yes, cost), without having to change anything else. Better still, assuming some users will still need laptops (in 100% of cases so far in my experience), the existing tools and processes work in exactly the same way for laptops, desktops and VDI VM’s. Electronic software distribution tools, such as SCCM or LANDESK do not even recognize that the some of the Win 7 desktops it ‘sees’ are virtual machines (they are full Windows 7 desktops after all), whilst others are physical desktops or laptops.  

Having given users what they want, the IT department then has time to go and fix those underlying process issues in a controlled and systematic way. These changes are often more complex (and so costly to do) than building the VDI environment itself. 

It is a huge volume of change, all at the same time, which kills many IT projects. This is especially acute in desktop virtualisation projects that try to do too much, too soon.

This is the key reason why application virtualisation is not ‘king of the hill’ in the desktop virtualisation world – and in my opinion, never will be.  Application virtualisation forces the organisation to go  ’all in’ on a change process that can represents a significant effort, even if they only ever intend to virtualize a small controlled set of applications. 

I was with a public sector client recently who are 20 months into their desktop virtualisation project (a mix of application virtualisation and RDSH’s), but only now have enough of their core applications packaged to begin their user roll-out. The client sponsor, a very experienced CIO (who inherited the project), commented to me that in his experience “the people who are doing desktop virtualisation now are all doing VDI”.

2. Security

My perspective on security is that it largely comes down to human behaviour.

Moving the desktop OS into the data centre, without doubt makes it easier to update and patch the software image – and importantly – easier to guarantee that each desktop has received the update (VM’s are always accessible, never leave the corporate – so updates always work, to every machine). It should be remembered that VDI VM’s are still Windows machines and still have to be managed, so you are not necessarily managing less images – it is just much easier to manage the ones you have.

A client in a large financial institution recently recounted that his last 5 security breaches had been caused by laptops that had ‘brought in’ some form of mal-ware after being used externally, because they had not been on the ‘home’ network recently and so had not received the latest security updates .  

This form of breach just can’t happen in a VDI environment – the VM desktop never leaves the network, so it always receives the latest update, as soon as it is released, and it is far less likely to be ‘attacked’ (never say never, of course) because it is hosted inside the corporate security zone.

From a human behavior perspective, because the VM image is completely separated from the end-point device there is no data retained on that end-point. This removes the possibility for bad luck and bad judgment to put your corporate data at risk.

I remember the celebrated case of an RAF officer in the 1990’s quite literally having the ‘Plans for the Gulf War’ stolen, as they were on a laptop in his car. That career limiting (ending?) event is simply impossible in a VDI environment – the device might still be stolen, but no data will be lost with the device.

3. People can use whatever device they want 

A big driver for VDI has been the desire to bring and use an iPad at work. It is, of course, equally true for almost every other type of device. The IT department literally no longer cares which device people use – they are all ‘untrusted’ devices that the IT department neither owns nor manages.

This works particularly well for contractors – several of our corporate clients make regular use of consultants and their default method of granting these folks access to their environment is to issue them with a laptop. This is very expensive for the organisation (even if occasionally get them back at end of the contract) and a real pain (literally) for the consultants themselves, as they now have TWO laptops in their bag (because they almost always carry their own laptop too).

In a VDI environment, the organisation simply issues the consultant a URL, username and password, allowing them to access their desktop from their own ‘untrusted’ laptop.

…and they can install what they like – on the local device, of course, not their VM image (although possible, still generally a very bad idea) as the local device is theirs and (see above) the IT department no longer cares what they do with it.

4. Instant ‘snow storm’ solutions

This is a great real world feature of VDI – the ability of people to log-on from anywhere, from any device means that everyone can suddenly change location and log-on their desktop with no pre-planning required.

So when the roads are suddenly impassable and no-one can get to the office, the business impact can be dramatically reduced.

This benefit can also be extended into all the various disaster recovery solutions and provision. For instance, several of our clients pay substantial sums for the provision DR office space, with PC’s pre-loaded with their image, sitting waiting for an event that (hopefully) never happens. VDI offers the opportunity to remove, or at least dramatically reduce, these costs.

Posted in Uncategorized | Leave a comment

Will Microsoft Licensing stop the adoption of Hosted Virtual Desktops?

Image

The latest ‘buzz word’ (well, OK, phrase) doing the technical conference circuit this year is that we have entered the “post PC era“. This conjures up images of lots of happy corporate users going about their daily work without any form of Windows desktop to access their applications and data, using <delete as applicable>.. iPad, MacBook, tablet computer, a home PC etc…..(it’s complete piffle, of course, for several years at least, because of all the legacy applications being run by businesses today that require a Windows environment. Until EVERY application in a business is somehow virtualised, executes in the browser or is made web accessible, then you’ll be needing that ‘ol skool Desktop).

To illustrate the point, from the ‘amazing-but-true’ department, I met a client last week still running a Windows 3.1 application …. I am not anticipating them moving that particular application into HTML5 anytime soon. 

So, given all the very real and proven benefits of running a hosted virtual desktop, and how much users love it for the most part (whether in your own data-centre = VDI, or someone else’s = DaaS), Microsoft licensing is often perceived as – and where ‘perception is reality’ actually is – a barrier to adoption for many organisations.

Wha’s up?

The almost impenetrable world of MS Licensing for hosted virtual desktops is a very real deterrent, as much for its complexity and convoluted logic, as for its costs.   

The root cause of the confusion amongst both customers and MS’s own re-sellers is the decision to continue to license Windows by device rather than by users. This is a quaint throw-back to the days when ‘Desktop’ literally meant a box on or under your desk and nothing else. This approach has thrown up some truly bizarre scenarios (for example: bring your own iPad to work and legally access your Company hosted desktop from the Starbucks across the street under your company’s SA agreement, but then bring the same device into the office and access the same desktop and you suddenly need an additional VDA license).

Rather than fix the root cause here and license the user Microsoft, hilariously quoting ‘customer benefits’ that they are seemingly alone in being able to identify, have invented yet another license called the Companion Device License (you guessed it, customer pays again!, for the same desktop). 

The implications of this are enough to give IT managers sleepless nights – instead of enjoying the benefits of users securely accessing their hosted desktops from any device, from anywhere, suddenly the IT department is faced with a mountain of work simply to track which user is accessing their desktop from which device, and from where, to remain in compliance … for devices they probably don’t manage or even own. Houston, we have a problem…

Give up now then?

… err, NO actually. In the real world Companies will buy the correct number and type of licenses for what they believe their users will actually do and the devices they think their users will use to access their virtual desktops. Of course Microsoft would rather you just forgot the whole thing and went back to PC’s – but the benefits of hosted desktops are just too good to miss for both users and Companies.

As for the cost – well, it is an additional cost that can be an unpleasant surprise for the unprepared. The good news, however, is that in my recent client experience there is enough cash benefit available in most business cases to justify the investment (and then some), even with these costs taken into account. 

Posted in Uncategorized | Leave a comment

Can VDI help the IT department survive ‘Bring your own Device’ (BYOD) in the workplace?

 

 

The advent of a multitude of very cool ‘smart’ devices is a nightmare for the hard pressed IT department in lots of organisations today. Regardless of the edicts of the security manager,  people are increasingly bringing their devices to work and using them (can’t bear to be without your MacBook Pro?)

Users are increasingly no longer happy to work on a 5-year-old laptop, running an OS that is 1 or even 2 iterations old, with ‘clunky’ performance – whilst sitting idle at home is the latest offering from <insert name of your favourite hardware vendor here>, which is ‘banned’ from the workplace for security reasons.

So, what if you could bring your device to work, connect to the corporate network securely in one window and ‘toggle’ to your own ‘personal’ window, when you need to do a spot of internet shopping (in your lunch hour only, of course) or book that hotel in the country you have been promising your partner for months?

What if there were no connection between the corporate network and the hard-drive of your (admittedly very snazzy) device, just in case you have let those anti-virus definitions lapse?

What if you could not download data to your device at all (unless the IT department specifically allowed you to do so)?

What if all you needed to do to perform this magic was an RDP agent for your <MAC, iPad, iPhone, Android device…>  (or if it’s a Windows device nothing at all, as it was already built-in)?

…but wait, doesn’t all this already exist?  

Luckily for the aforementioned ‘hard pressed IT department’, it does. VDI offers exactly this, without requiring the IT department to invent and then manage a whole new set of business processes to secure and manage devices that the company doesn’t actually own.

As an added bonus, that very same VDI desktop suddenly offers remote and home workers easy and secure access to their corporate desktop, over the internet, from any device.

A client recently commented to me that the main thing driving user adoption of VDI was that it no longer required people to use the existing slow and cumbersome  SSL VPN solution to get remote access from home.

I am an un-ashamed fan of VDI – I just love the flexibility of being able to access my desktop lots of different devices (none of which I have to carry, unless I want to). Of all the great features of VDI, this is one of the best from a user perspective – but I strongly believe that getting lots of end-point devices ‘off the books’ of the company makes it one of the best features for the IT department too.

But doesn’t this all cost LOTS of money?

As the price of PC’s has come down, so too has the cost of VDI. The new generation of VDI brokers are remarkably ‘light’ in their use of infrastructure (server, storage etc,.) and the good ones avoid expensive SQL licenses and so forth. This means the business case for adoption is starting to add up at very low user numbers, in spite of Microsoft’s very confused and complex licensing (which just got worse, not better, with the Companion Device License – of which more another day).

The cost of building a VDI environment can be very close to the cost of buying laptops, whilst being much cheaper to support and run, resulting in a very healthy saving in overall total cost of ownership.

So, happy users, happy IT department, happy CFO … are laptop manufacturers the only losers here? Perhaps this explains Dell’s recent big moves into the Cloud with their acquisition of Wyse and their new alliance with Desktone, amongst others. It looks like a smart move from here….

Posted in Uncategorized | Leave a comment

Molten Technologies in the press

London, UK – 22 November 2011Molten Technologies is proud to announce a contract with Thames Water’s lead service integrator and Desktone for the provision of virtual desktops to Thames Water. Using its desktops-as-a-service (DaaS) platform, Molten will support a substantial portion of Thames Water’s third party and internal staff desktop.

This has resulted in several press articles:

  • CIO – Thames Water turn on VDI from Molten Technologies
  • Computer Weekly – Thames Water uses refresh cycle to pilot virtual desktops
  • CRN – Molten to melt away Enterprise VDI woes
  • Techworld – Thames Water deploys VDI from Molten Technologies
  • The Virtualization Practice – DaaS Throws off it’s waterwings

Watch this space, we will keep you informed.

Posted in Cloud, DaaS, Desktop as a Service, desktop virtualisation, End User Computing Strategy, VDI, virtual desktops | Leave a comment

VDI counter arguments – it won’t work with iTunes

Corporate iTunesAs CEO of Molten Technologies, an independent virtual desktop specialist, I often find myself enthusing about VDI, especially as a service, vs the traditional fat-client PC model and I have heard every push-back in the business (and some that ought not to be). In this series of articles, I will expose the most common and a few of my favourite rarer ones. This particular article is an example of the latter.

I know this is hard to believe, but I was talking to a senior IT manager from a large multi-national enterprise who appeared genuinely concerned about how his staff (and possibly himself) would handle iTunes if we moved them onto virtual desktops. You may laugh (I actually did as I recall), but the concept that the corporate laptop is a perk to be used for personal as well as business ends is more widespread than I had realized. In my opinion, however, it is out of date and has no place in today’s enterprise IT strategy to be replaced by a more subtle approach.

First, let’s deal with the raw technical facts. iTunes can be made to run on a virtual desktop and it can synchronize with local devices. I know this, because I have tried it (for purely journalistic purposes, you understand). Your administrator needs to have set you up in such way that your session will map to local drives and devices. This is precisely the opposite of how I usually recommend Enterprises set up their virtual desktops, but this is about “could I” for a start, we will come to “should I” in a moment. Performance is interesting when synchronizing with a local device (like my iPhone) because it is now happening over the internet, but it works as long as you have a decent internet connection (to be fair, I am on fibre and have not tried it over lesser connections).

Second, (in the style of Chandler Bing from Friends) “oh-my-god why?”. I get the potential for good technology to be an attractor for new staff and a retention tool for existing staff, however, there is an alternative to opening up your devices and networks to this sort of abuse.

A well set-up, performant, virtual desktop is a perfect way to segregate work from personal life without having to carry two devices. Inside the virtual desktop you can allow secure, controlled access to your corporate network and outside can be an unmanaged device of your choice.

VDI therefore supports a “Bring Your Own” approach, allowing staff to use their Macs, iPads and Androids at work. So if you really must have access to iTunes at work, then you can run it on your own device on the guest network (and I will leave your manager to worry about whether you are doing so on his time or your own).

This is a step in the right direction for many technologically sophisticated employees because they can now use their beloved personal computers (and I mean PC in the literal sense) in a work environment while neatly keeping their personal and business life separate (including associated data-loss risks and viruses). If you want to go a step further to attract the right talent, you could offer a stipend to staff to encourage them to buy their own or even buy certain groups a trendy tablet device. In either case, my strong advice would be to leave ownership of the resulting device with the employee and responsibility for management with it. If you give access to their work life through a virtual desktop rather than locally on the device then you can leave the device both unmanaged and untrusted.

The model is changing from late nineties and the noughties (I love that term) where the best technology in the house came from the employer as a perk-laptop through a frustrating phase where we all had brilliant tech at home and work was still making us use clunky, heavy laptops into a brave new world where we can carry whatever technology we like (or even carry none and use what comes to hand) and still have secure, performant and convenient access to our working environment wherever we go.

Posted in Cloud, DaaS, Desktop as a Service, desktop virtualisation, End User Computing Strategy, Uncategorized, VDI, virtual desktops | Leave a comment

VDI counter arguments – It’s too expensive

As CEO of Molten Technologies, an independent virtual desktop specialist, I often find myself enthusing about VDI, especially as a service, vs the traditional fat-client PC model and I have heard every push-back in the business (and some that ought not to be). In this series of articles, I will expose the most common and a few of my favourite rarer ones.

 Traditionally VDI has been expensive to set up and there remains a tendency to compare the cost of a big pile of laptops with the cost of a virtual desktop infrastructure. As we all know, the PC market has become highly commoditised and the price of laptops, particularly notebooks, on the high street has dropped like the proverbial stone. So what does the comparison look like today and is it a fair one?

Let’s start with what the comparison looks like today, before diving back in to the thrashing waters of debate about total cost of ownership and whether virtual desktops are cheaper to run than physical ones. Molten Technologies regularly responds to tenders for virtual desktop infrastructure (VDI) and builds bespoke VDI infrastructure for our clients, so we have an extremely up-to-date view of what it costs to build such infrastructure (because otherwise we would lose a lot of business or money).

For example, we recently quoted a client for a 2000 seat VDI infrastructure build, including data centre hardware, server and build effort at £620k. This was for “true” VDI with a dedicated virtual machine for each user, so not terminal services or application virtualisation, but the full-fat stuff. This equates to £310 per “bare metal” machine (I.E. without OS and application licensing). In comparison to the typical corporate laptop, this is clearly cheaper and was a big part of the business case for this client to provide virtual desktops for its contractors rather than physical ones. In this case, there were no thin client devices to buy, because the client expects the contractors to bring their own devices and has the virtual desktops locked down to prevent access to local drives and peripherals (other than keyboard and mouse), cut and paste functionality as well as second factor authentication to guard against key loggers on the unmanaged devices.

You may argue that we should include thin clients for a fairer comparison, but frankly even if you add £150 (real figure from a recent client project) you get a combined up-front cost of £460. This is still competitive with what most corporates pay for their fat-client machines.

So is VDI “too expensive” to buy? Only if you are doing it wrong, is my conclusion. Sure, you can buy a fat-client for less than £460 on the high-street, but you are likely to spend more if you are getting the sort of processing power that most corporates require.

What about too expensive to run? Surely a virtual infrastructure costs more in licensing, networking and general hassle. Well, there are some swings and roundabouts and the comparison between them will vary depending on the user group and the organisation in question, but in general our experience is that virtual desktops ought to be cheaper to run than physical ones.

Areas of run-cost increase include:-

  • Virtualisation licensing (these costs can vary dramatically, so it is worth seeking out the best solution for your situation rather than defaulting to the same organisation you used for server virtualisation just because your IT team know them. You can smile about this one, but it happens quite a lot).
  • Increased Operating System and Application licensing (this is usually code for Microsoft being confusing and even down-right unhelpful if you want to run your MS software in a virtual environment. The rule of thumb is that if you have software assurance you are OK, otherwise they will find a way to tax you).
  • Network availability (virtual desktops typically require less bandwidth than fat-clients, but are more reliant on the network, so there may be some remediation in terms of reliability of the network)
  • Support for your virtual desktop infrastructure (in the same client quote mentioned above, the VDI support cost was £36 per user per year. This buys you the virtual equivalent of PC hardware support as it is in your dreams, that is 99.9% available with a hot standby to ensure that figure if there is a hardware failure).

 Areas of cost-reduction include:-

  • Provisioning effort is dramatically reduced as there is no physical machine to roll-out and hand-over to the user. The provisioning process is now simple matter of using an online toolset to create a fresh desktop from one of your existing gold patterns, and emailing the details to the user or their team leader.
  • Decommissioning is now a relatively simple button-push process rather than a very hit and miss physical collection (especially if we are talking about contractors).
  • At elbow support and repair of damaged devices can be reduced by having some spares in each office and an understanding in the users or managers that they simply swap out the device if there is a problem. The thin-clients are not personalised, so any one will work just as well as another for each user, without the need to copy any settings, applications or files. At elbow support can be reduced to zero if the user groups is contractors who are required to bring their own PCs because they remain responsible for their hardware.
  • Help desk support is typically reduced for a good VDI installation. This is a hotly debated topic and there are those with a different point of view, but our experience is that Windows simply runs more reliably on a hypervisor than it does on PC hardware. I am not absolutely certain why this is, but it seems likely that all the “helpful” add-ons and functions that the hardware manufacturers provide to differentiate their products affect stability and so do some of the functions associated with the hardware form-factor (e.g. hibernation on a laptop). Whatever the cause, the facts from the field suggest that Windows and applications running under Windows give rise to fewer helpdesk calls by about 40%.

Consideration of the expense of virtual desktops would not be complete without a mention of “cloud” desktops or hosted virtual desktops which offer a rental-style model, increasing operational cost but minimising capital expenditure to very small numbers indeed. This can offer a lower barrier to entry and greater flexibility and is a service that Molten Technologies offers (so I know a fair bit about that model too). This model is worth considering and will certainly offer a cheaper solution at lower volumes. Our website proudly offers a virtual hosted desktop for £1 per day, which is a very good deal between twenty and two hundred desktops, but when you get into the thousands, you should be looking for a discount on that in order for it compete with the sorts of numbers I have talked about above.

I would always recommend a full cost of ownership (TCO) comparison for each user-group before moving from thick-client to virtual desktop or any other substantive shift and there will be user-groups for whom this does not make sense, either economically or practically. However, it is clear that buying virtual desktop infrastructure (VDI) is substantially cheaper than it used to be and that it can and should be cheaper to run than a PC estate. If you don’t believe me, then give me a call, Molten Technologies will quote a custom-build virtual desktop infrastructure (no charge for the quotation) and do the TCO comparison for your organisation.

Posted in Cloud, DaaS, Desktop as a Service, desktop virtualisation, End User Computing Strategy, Uncategorized, VDI, virtual desktops | Leave a comment

VDI counter arguments – User Experience “sucks”

Help your users to love their virtual desktops

As CEO of Molten Technologies, an independent virtual desktop specialist, I often find myself enthusing about VDI, especially as a service, vs the traditional fat-client PC model and I have heard every push-back in the business (and some that ought not to be). In this series of articles, I will expose the most common and a few of my favourite rarer ones.

Here is a verbatim quote from a potential client, “VDI has limited use here because the user-experience sucks”. This is not, I hasten to add, a client who had experienced Molten Technologies’ service, but it is not an uncommon VDI experience. I would love to be able to say “this used to be true, but we have all left that far behind us”. The truth is closer to “it no longer needs to be true” with an unspoken dollop of “but it still might be unless you engage the right skills in setting up your VDI capability”.

There is absolutely no reason why a well set-up VDI service can’t be faster and a more pleasant user experience than an equivalent PC. Indeed, because of the natural pooling of resource (something that the PC world has been trying to emulate with limited success for years) and because the OS will be running in a data centre where it should have much better connectivity than the average office, it is likely to be cost-effective to make it even more performant. I use our own hosted virtual desktop solution all the time and it opens Outlook, for example, in three seconds, as compared to about a minute on my laptop (see our YouTube demonstration here).

First of all, let me be clear that I am not going to push a particular supplier’s solution here. I make no secret of the fact that I have my favourites, but in this instance there are a wide range of choices that will work fine. However, you will need to follow some basic rules and avoid some pitfalls.

1 – IOPS, IOPS and more IOPS

“IOPS” stands for “Input/output Operations Per Second” and refers to the speed at which storage can respond by writing to disc (“Inputs”) and retrieving from disc (“outputs”). The PC has a local hard-drive dedicated to the machine and over the years PC operating systems (you know who I am talking about) have become very “chatty” with their local drive because they can get away with it without a performance impact (some might call this laziness on the part of the OS designers). Translate that same behaviour into a server architecture in the data centre, where the storage is likely to be on the network rather than local and where the hardware isn’t designed to deal with the sort of “lazy chattiness” of a PC and your IOPS will become an instant bottleneck. There are many ways to optimize your IOPS, starting with the use of specialist storage that can hold the most common calls cached in local memory. This works because the OS might be chatty with the disc, but it’s “conversation” is tedious and repetitive, so most of it never really needs to go to the storage at all, because it is the same as the last request. It may also make sense to use storage that can “de-duplicate” across the desktops. This means that if two virtual machines both hold an identical copy of “Word” the application will be stored once with a pointer to both machines. Given the size of the OS and the applications, most corporate desktop estates can save about 80% storage on the desktop image, because they are all storing the same thing.

2 – Image is everything

Your average desktop image and its management processes are intended for a PC, possibly a laptop PC, and there are a number of areas where it can be optimized for use in a virtual machine which remains permanently in a data centre. For example, the anti-virus can now be run at a server level rather than on each virtual machine, which is dramatically more efficient (if you are going to run it on each machine, make sure it is staggered and at night). Patching can also be done at night, or whenever the machines are not being used, rather than at the point the user first logs-in; ditto with the virus scans. Think about the fact that your virtual machine is always on, always connected and never leaves your corporate network and you will find a number of opportunities if you think hard enough and with a sufficiently open mind.

3 – Know your customer

This sounds obvious, but is remarkably often omitted. Is the end-user going to need video and two-way audio? If so, then make sure you are using a remoting protocol that can cope with these demands. Old-fashioned RDP cannot do two-way at all and makes video look dreadful on all but the fastest networks. RDP 7 is a load better and there are a host of other options out there. Consider how to best balance ubiquity (there is some version of RDP on almost everything out there from iPhones and Android to MacBooks and Windows NT), cost (some of the options can get expensive) and user need.

If the user is going to run very memory-hungry applications intensively then make sure they are allocated more “power”. This isn’t hard to get right and there are tools to help you monitor actual usage so that you can continue to optimize it over time.

4 – Networking

As you move your “PC” from the desk in front of you to the data centre, your networking requirements change significantly. This will provide you with some opportunities for increased performance (e.g. put it in the same DC as some big, data hungry application and it will run like a rocket for the first time ever). However, it may also create some challenges (e.g. if you moved your file-store into the office for better performance, it may now get worse, unless you move it back into the DC with the virtual machine). The end-users will typically require less bandwidth to their local machines than they would have done for in a thick-client environment, but may be more sensitive to latency.  For most office networks, this will not be a problem, but is different from what we are used to. For example, a strong 3G signal may give a better experience than a poorly set up WiFi.

These are just some examples, the bottom-line is that if you have the right skills and experience on hand to help you think through your virtual desktop implementation, it can be a performance dream. If you employ a generalist to “slap it in” it could become a nightmare. There is no reason for your user-experience to “suck” and there are opportunities that may enable you to make a performance improvement over thick-clients.

Posted in Cloud, DaaS, Desktop as a Service, desktop virtualisation, End User Computing Strategy, Uncategorized, VDI, virtual desktops | 38 Comments