Pivot3 has released its second generation of VDI appliance, looking to make Pivot3 vSTAC R2 simpler to implement and more scalable and to drive further savings than its predecessor. Headline features include utilising VMware View Storage Accelerator, allowing reduced hardware costs; performance increases with more memory and updated processors, allowing greater VM density; and networking options, enabling adoption by a wider SMB market.
We covered Pivot3′s first release about this time last year. If a week is a long time in politics, a year is a generation in IT. In that time, we’ve covered the fact that Pivot3 is not alone; there are¬† range of appliance makers are looking to simplify VDI adoption.
Pivot3 claims vSTAC VDI R2 can deliver each desktop for the eyebrow raising sum of $165. How are they doing that? Will the improvements help push VDI adoption? Have Pivot solved all of the key questions of VDI appliance makers?
Why do VDI Projects Stall?
Implementing virtual desktops promises a range of CTO/CFO attention-grabbing phrases: Administration savings; enhanced security and auditing; reduced disaster recovery costs; improved workforce mobility; increased user productivity; chuck¬† BYOD in, everyone likes a bit of that; faster acquisition assimilation; reduced remote office integration costs.
But we say again, and will keep saying until it isn’t true: desktop virtualisation is not the same as server virtualisation.
Yes, the process of assessing and planning a transition is similar. Very simply, you‚Ä¶
- ¬†¬†¬† Determine what it is that you need
- ¬†¬†¬† Measure what it is that you have
- ¬†¬†¬† Assess how it would be delivered in a virtual environment
- ¬†¬†¬† Implement and test a virtual environment
- ¬†¬†¬† Roll-out
- ¬†¬†¬† Monitor, re-measure, and re-assess as required.
There are a number of reasons that this process affects VDI projects. Two.. er, three.. actually four.. four stand out (what have the Romans ever done for us, eh?):
- Infrastructure Complexity ‚Äď The nature of desktop use is complex and varied. You rarely ‚Äėreduce‚Äô infrastructure with VDI; more likely, you increase it. Fundamentally, this is often because the core desktop OS being virtualised was not designed to run in a virtualised¬†environment. VDI projects typically require the introduction of SAN infrastructure and additional networking, as well as virtualisation software. This leads to a very high capital expenditure cost.
- Support Multiplicity¬†‚Äď¬†Given this additional hardware and virtualisation software, support costs can increase in order to support the SAN and changes in workspace and application delivery models. In addition, any outages in the back-end VDI infrastructure can affect a great number of people. Any potential savings from operational expenditures can take time to come through, and they are often difficult to realise, especially with smaller organisation sizes.
- A remote desktop is not a physical desktop¬†‚Äď¬†This might seem like a no-brainer, but there are assumptions made when creating a virtual desktop environment, especially when you are looking to drive down costs. For instance, the user will have a fixed application environment, or their application resource requirements will be standardised. The actual experience may be different (in terms of response time, and audio and/or video¬† display). It is not unusual to find a poor experience not directly related to the hardware running the VM; understanding that is key to resolving issues. Users’ ability to change the environment that they are working in may be reduced or removed.
- A strategic level view for VDI is vital for a successful deployment¬†‚Äď¬†Why? Because in understanding that how virtualised desktops will fit and deliver to the business strategy will shape both what is required and how those virtual desktops should be delivered, it answers the question of what part they will play in the overall enterprise desktop strategy/IT service delivery plans.
Not addressing these issues early in the project cycle is a fundamental problem for VDI projects.
What is new in Pivot3 vSTAC R2
This generation of the Pivot3 VDI appliance exploits View Storage Accelerator; the vSTAC VDI R2 appliance can therefore reduce the need to work read-caching hardware, without compromising VM disk performance.
- Two next-generation Intel¬ģ Xeon¬ģ E5-2630 processors with six cores each. Pivot3′s testing shows a this allows up to a 15% increase in VM density over the previous generation.
- Support for 1Gb Ethernet environments. The new generation has a configurable network speed configurability, maintaining support for 10GE and introducing a 1Gb Ethernet environment, increasing the number of environments where this appliance can be deployed.
- RAM options for 128/256/384GB. These options allow better support for demanding 64-bit desktop workloads
- SATA storage capacity of 12/24/36TB.¬†This allows for user data to be stored in the same array, with an optional NAS interface. In a VDI environment, fast access to file store is key for application performance and user experience. Storing user data on the same array will present the option of allowing branch office desktops to be managed and deployed using VMware’s newly acquired Mirage technology.
The appliance has been tested and certified for VMware View 5.1 and VMware vSphere 5.1. It has also been tested and certified for PCI-e graphics cards such as Teradici. And, it is possible to obtain a pre-packaged and configured P Cubed starter appliance that is certified as part of the VMware Rapid Desktop Program.
Excited? Intrigued? It does sound like a very interesting prospect.
Can VDI Appliances Deliver?
In discussing how VDI appliances simplified VDI adoption, we looked to draw attention to the fact that while VDI appliances can significantly reduce sizing and integration risk from a project, they are still not a complete end-to-end solution. There is no VDI solution, yet, that comes ready-made out-of-the-box. VDI appliance vendors have key target markets in education, local and state government, healthcare, and distributed financial and legal services, with user bases between 250-2,500 desktops. It is likely that the challenge of a remote desktop is going to fall foul of the need for customisation or differing user application resources.
VDI Appliances allow for reduction in complexity of environment. Pivot3′s new StormCatcher feature allows for disk resources to be utilised across appliances. Unified flash accelerators and improved memory, CPU, and networking options in Pivot3′s vSTAC increase density per device. Pivot3 has also worked hard to ensure that adding a new appliance has benefit to all desktops in the environment, rather than simply increasing the number of desktops. In terms of reducing complexity and cost, boxes are ticked.
Still, it would be useful to have recognised VDI industry standards on the testing and verification of VM numbers. Use of an independent testing methodology to give a baseline that could then be customised to give a value for an organisation is the way forward here. Assessment of user requirements and applications and testing of the process prior to migration is key to the success of a project. Indeed, a VDI implementation beyond a workgroup- or project-based environment still needs to be combined with tools to enable customisation of the user desktop environment and application stack and tools to report on user experience.
Appliance vendors (be they hardware or software) need to consider better integration with user profile/application installation tools and user experience monitoring: applications from the likes of RES Software, Centrix Software, Lakeside, Liquidware Labs, Norskale, Unidesk. Hardware limitations, such as IOPS, are often problems that have been “solved”, traditionally with more hardware and money. Here Pivot3 has looked to work smarter and cheaper.
The Pivot3 vSTAC R2 appliance has looked to solve a number of density and performance issues and improve on the previous revision. Yet poor VDI user experience often goes beyond physical hardware as a solution. Poor user experience is not always simply about an issue at log-on/off, but a wider user context of access to data and ease of installing, changing, and controlling applications. From an administration point of view, it is not simply a question of whether the hardware is available, but also of the responsiveness of applications throughout a user session.
Pivot3 continues to drive the hardware cost and complexity of VDI down. The headline $165 / desktop will be compelling. The performance assurance and dynamic expansion in R2 will likely solve many administrative headaches. The key for any turnkey VDI solution remains ensuring that the assessment and testing prior to migration is undertaken properly, and that once it is installed, there is effective management to deliver on the wider user experience.