Perhaps it’s just me but wasn’t the brave new world of cloud computing going to solve world hunger? Well, that’s the way I understood it when I attended all those technical and commercial seminars about cloud resources around the turn of the last decade.
Actually as far back as 2008, through The 10 Laws of Cloudonomics some thought leadership was available. Joe Weinman, then Strategic Solutions Sales VP for AT&T Global Business Services, created laws that still are the foundation for the economics of Cloud Computing.
Cloudonomics Law #1: Utility services cost less even though they cost more. Although utilities cost more when they are used, they cost nothing when they are not. Consequently, customers save money by replacing fixed infrastructure with Clouds when workloads are spiky, specifically when the peak-to-average ratio is greater than the utility premium.Joe Weinman
In all fairness, nobody has a crystal ball so I’m sure that we all hoped that industry experts were correct in their ideas and beliefs at that time. The reality, as it turned out, is far from what we were being promised.
Only pay for the cloud resources you use
For me, the biggest misconception was the notion that you could simply eliminate all the capital expense of your massively costly data centre equipment and perhaps even the data centres themselves; resulting in transitioning entirely to an operational expenditure (OPEX) cost model for cloud infrastructure and paying only when you used it.
Anyone that has tried to do this knows this is nowhere near the reality of cloud migration. In fact, many of the companies that have attempted this found themselves having to move back to a traditional infrastructure. Many stating that the costs of cloud are far too high and certainly not what they had been led to believe.
This reality has also been recognised by Microsoft (and other cloud vendors) resulting in the introduction of reserved cloud resources. Sure, if anyone commits to a year or three-year term they are certainly likely to get the cloud resources cheaper than they would if they stay with pay as you go but many applications like desktop virtualisation, for instance, do not lend themselves at all to reserved pricing. This is mainly due to the unpredictable nature of workloads like AVD and Citrix VDI. This has made running virtual desktops in the cloud practically impossible from a commercial perspective.
Until now. Essentially that is the reason why we founded Cycloud.
Cycloud Provisioning Services
Cycloud is the technology that has been specifically designed to reduce the cost of running virtual desktops like AVD and Citrix VDI on public cloud platforms, in this case, Azure. How does Cycloud help? Well, the technology is focused on provisioning (or creating) cloud instances only when they are needed and deprovisioning (completely removing) them when they are not.
Simple, right? Well not really.
For a start, how does anyone know when virtual desktops are needed? Most admins know the basic working patterns of their users, but they have no idea how to match the provisioning of their desktops and applications to when users actually want to use them and that is usually why they simply create what desktops they think they need and hope that they are right. Invariably they are not, usually purchasing more than they actually need, inevitably experiencing the shock and disappointment of receiving cloud bills reflecting the inaccuracy of guesswork.
Aren’t there solutions that power down the desktops that can save money? Yes, but they do not completely deprovision the desktops leaving the costly Azure storage charges existing in the cloud and continuing to consume substantial costs even though the desktops are not using it.
Cycloud completely removes all aspects of the virtual desktop so you can get the most comprehensive savings and because Cycloud scheduling allows for the repeating of schedules this means that AVD and Citrix virtual desktops are created the next day and the cycle repeats to save costs.
Scheduled Provisioning of Desktops
Scheduling the provisioning of desktop resources is one way of reducing cloud costs and it is fair to say that most of the time admins know when users start work. An example would be that users start work at 08:00 in the morning and they will need desktops to work with. Cycloud has advanced scheduling capability that can provision from one-to-many hundreds of virtual desktops at any time of the day and make them available to users.
Cycloud’s advanced scheduler is also able to deprovision virtual desktops at the end of the day when users no longer need them.
Robotic Desktop Provisioning
Cycloud technology can also sense when more desktops are needed and when desktops are not being fully utilised and need to be deprovisioned. This provides a means of autoscaling desktops to accurately match the business requirement.
Autoscaling up (or automatically adding desktops) is achieved by looking at the performance of the desktops and if it is decided that they are not performing inside acceptable limits then more desktops are provisioned. Autoscaling down (or automatically removing desktops) is achieved by looking at the user count on the machines and deciding if for instance, no users are using them then they should be removed.
Setting up scheduled and robotic provisioning correctly allows companies to ensure that they are only using desktops when their users need them and this ensures that the costs are in line with actual requirements.
Cycloud has built on the expertise and experience of our engineers and has been specifically designed to help companies reduce the cost of their virtual desktop provisioning in public clouds. Focussed on Microsoft, Cycloud has become one of the leading Azure cost optimisation solutions.
We believe that with Cycloud any desktop cloud deployments can be as cost-effective as traditional deployments but with the added benefits and flexibilities of public cloud. Customers can see the extent of their costs with the Azure pricing calculator and with Cycloud’s advanced scheduling acting as your Azure cost optimiser, Azure virtual machine cost savings of well over 65% can be easily achieved.
To see how much you could be saving, try our Cost Savings Calculator now.