Desktop Virtualization On Your Terms – Flexibility and Choice with Architectures That Fit
I recently had the opportunity to host several customers in a roundtable discussion, exploring their experiences in deploying desktop and application virtualization, the challenges encountered, and the benefits they’ve reaped. It was an engaging dialog with organizations spanning mid-market, enterprise to large service provider environments deploying either Citrix XenDesktop or VMware Horizon View desktop virtualization software. In case you missed it, you can check out the event here. I mention this because it provides a valuable backdrop to some important news Cisco is sharing today, centered on helping IT organizations (like those I met with) more quickly achieve success in VDI.
Over the last few years, Cisco UCS has rapidly established itself as a leader among competitors with a much longer history in the server marketplace. Why is that? If you talk with anyone who’s implemented UCS in their data center, they’ll instantly tell you about the operational streamlining and simplification that UCS Service Profile Templates offer, the value of a unified data center fabric for LAN and SAN, and the performance derived from a platform that was purpose built for highly scalable, virtualized environments.
It should be no surprise then, that when organizations evaluate their options for server infrastructure to host virtual desktop workloads, the same UCS core value proposition extends nicely to desktop virtualization – the benefits of which are multiplied, in fact, given that virtual desktops can consume infrastructure resources and capacity in unique ways compared to other mission critical enterprise applications. We’ve therefore seen great response from our customers (as demonstrated in our webinar/panel discussion) when it comes to the fitness of UCS in hosting virtual desktops.
What we’ve come to find through our customer’s experiences, is that the vendor marketplace has traditionally taken a one-size-fits-all mentality around VDI architectures that either forces organizations to overspend CAPEX on approaches that are tuned for much larger environments, or wrestle with an economized approach that results in poor desktop user experience. Clearly, there’s a spectrum of IT implementation use cases that apply, when we’re talking VDI. Persistent desktops vs. floating, SAN in place vs. greenfield, one-hundred seats vs. tens of thousands, etc. so one size will never adequately fit all!
For this very reason, we’re expanding our portfolio of desktop virtualization solution architectures, along with the ecosystem of technology partners who are helping us accelerate the path to VDI success for environments of all sizes. While Cisco enjoys a strategic relationship with NetApp and EMC, we’re now offering desktop virtualization solutions that also include technologies from partners such as Nimble Storage, Nexenta, Atlantis Computing, Fusion-io, Tegile and others in process.
With these partners’ technologies come new capabilities that exploit key trends in the VDI and data center marketplace, including the proliferation of flash-based storage solutions, and appliance based approaches that mitigate the need for embedded SAN infrastructure and expertise (especially in smaller environments). Additionally, unlike our competitors who are narrowly focused on their own storage portfolio, Cisco can offer our customers the flexibility and choice they desire in selecting the storage technology and solution for VDI, that best fits their environment.
I encourage you to learn more about this exciting new portfolio of architectures by checking out the assets below.
Today Cisco is introducing an expanded architectural portfolio and partner ecosystem in support of our successful desktop virtualization solution built on Cisco Unified Computing System (UCS). Cisco UCS market traction has been phenomenal over the last 3 years. In fact, desktop virtualization has been one of the top workloads deployed on UCS as IT organizations apply the benefits of our stateless, simplified operations model, expansive I/O, and scalable performance to desktop workloads in the data center. Combined with unique product integration and the software eco-system partners such as VMware, Citrix and Microsoft, Cisco has delivered a number of reference designs with our strategic storage partners such as EMC and NetApp. Typically, these architectures were based on designs that easily scale from supporting a few hundred virtual desktops to thousands of desktops.
We have seen an inflection point with the perfect storm of the evolution of storage options, desktop software maturity, and data center architectures. One of the important changes in the storage market is the emergence of flash storage to address performance problems.
Taking advantage of enhanced UCS features and expanding the eco-system of storage partners including Atlantis Computing, Fusion-io, LSI, Nexenta, Nimble Storage and Tegile, Cisco is defining a broader portfolio of data center architectures for delivering desktop virtualization solutions – on-board architecture, simplified architecture and scalable architecture. “Converged” or “Unified” infrastructure stacks such as FlexPod and vBlock have, and will continue to be another successful option for desktop delivery infrastructure. Let me walk you through each of these architectural approaches.
I had the opportunity to chat with David Yen a few days ago on a number of topics--one of the things he touched on was how he sees the data center evolving. Now seeing as David is the Senior Vice President and General Manager of our Data Center Group, there are more than just idle musings. Here is a snippet of our conversation:
Omar Sultan: David, you talk about the evolution to an application-defined fabric--from a practical perspective, what does that mean to our customers?
David Yen: We are seeing a shift from a static, IT-controlled environment to a highly dynamic, user-driven environment. The net effect is to bring IT and the business closer together so that is good, but there are some practicalities that need to be addressed in the process. Amon the things we are focused on is making IT easier to consume for app owners and making this dynamic new environment easier to manage for IT.
OS: So, what are we doing to help customers make this transition?
DY: Well, we have been giving them the tools to prepare for this on-demand world for over five years now--our entire Unified DC portfolio—Unified Fabric, Unified Computing and Unified Management --is built around making data center resources flexible and more responsive to quickly changing user demands.
Unified Fabric allows customers to quickly and easily provision network and storage access wherever and whenever they need it. Similarly, UCS Service Profiles allow a UCS server to quickly and automatically adapt to the specific needs of a new workload. We have an entire portfolio of complimentary VM-networking technologies that then ensure there is consistency between the physical and virtual environments. Finally, Unified Management orchestrates, automates, and puts the infrastructure at your fingertips. Today, you can completely configure infrastructure for your apps with a few mouse-clicks. And with Cisco ONE, we are now adding the programmatic interfaces so apps and other systems will be able to directly configure their infrastructure for themselves.
While we have been doing this for a while now, it seems some companies are just catching-up. Recently, we saw a competitor claim leadership in the data center, but if you closely examine their claims, they announced things we have been shipping for a while: cloud-optimized architecture: check, on-demand resources: check, orchestration and management tools: check, L2 Multi-Path: check. Its actually more interesting to note what’s missing—things like network and compute integration, hybrid cloud capabilities, service chaining and multi-hypervisor support. Speeds and feeds are always important, but if that’s all you can talk about, then you are not going to be relevant to today’s conversation.
OS: Where are we going next with the data center fabric?
DY: Looking ahead, there are a couple of areas we will look to address. First of all, while we know that customers are aggressively moving to VM and cloud-based workloads, there is going to be a significant transition period and most enterprise data centers will remain a mix of physical, virtual and cloud workloads and we want to give customers a more comprehensive approach to dealing with this. At the end of the day, the data center should be able to deal with all types of workloads as equal citizens. We don't have that today in the industry--we have to resort to gateways and other mechanisms to span across physical, virtual and cloud domains--while that's OK in the interim, its problematic in the long-term.
The other area we will address is increasing operational simplicity. In this dynamic environment, it is neither feasible nor desirable for network operations to be involved in every config change. Ultimately we need to be able to do things at machine speed. You have seen some initial steps in that direction with the Nexus 1000V and its hypervisor integration or new technologies like Power-On Auto Provisioning. Our work with Cisco Open Network Environment has given us the tools and mechanisms to open networks up to facilitate these machine-to-machine or application-to-machine conversations through APIs like onePK and REST and through support of SDN controllers and agents like OpenFlow.
OS: David, why should customers remain confident about Cisco's vision?
DY: Betting on Cisco is not an act of faith--time and again, we have lead market transitions and delivered the technologies customers need to take advantage of those transitions. We are still, by far, the preferred networking choice, even in the most demanding environments like Massively Scalable DCs, where we are in production for 9 or 10 of the largest providers. We have more than 40,000 NX-OS customers and over 11 million 10GbE ports out there. This gives us unmatched insight into what customers are actually doing and where they are going with their networks. Similarly, we will be delivering VM network solutions across all four major hypervisors, which gives us unmatched breadth of experience in that space. Central to this longevity is avoiding technical blinders. UCS was a great example of our willingness to start off with customer needs in mind. Everything was on the table and that led us to breakthroughs like a brand new operations model based on service profiles. This willingness to risk and lead has translated into to remarkable growth in a very demanding market against a number of capable and entrenched competitors.
As I look at the competition, I see two hurdles they must clear. The first is simply one of simple experience. Its one thing to have a theoretical understanding of a technology and its quite another thing to have actually built and supported it. We have being shipping our Nexus 1000V virtual switch for four years now--we are into third generation functionality like hybrid cloud transport, cloud-based routing services, service chaining and multi-hypervisor support. Compare this to companies that are just getting around to shipping their first virtual switch and will still be working through first generation features and problems.
The second hurdle is a matter of getting caught up in a technical agenda instead of focusing on the customer's agenda. Software in networking is all the rage right now, for some very good reasons, but you see companies that want to shift all the network functionality into the software because that suits the narrative they want to tell. Now you and I both know, there are some things that absolutely are better handled in software, but, by the same token, there are things are better handled in hardware. We have control over both and that gives us the freedom to put functions where they are best handled. We think that will always give us an advantage over companies that are locked into a particular narrative and must make compromises to support that story.
To hear more from David, and trust me, he has some interesting and entertaining things to say, check out his Solution Keynote on Monday, June 24 at CiscoLive in Orlando.
A story came out today showcasing the platform built to power Xerox Managed Print Services (MPS), and it's hard to describe better example of how Cisco's data center technology comes together to help unlock the full potential of cloud computing. In the lead-up to this release and the webcast that we're airing this Thursday, I had the pleasure of working with Tom Force who leads up the architecture team at Xerox that built the MPS cloud. What I heard him describe illustrates some of the fundamental differences in UCS that come into play for cloud builders:
Fabric-centric design. MPS is a cloud based service and hosts over a hundred applications. Many of these are multi-tier apps and they benefit directly from the fact that every server in a UCS environment is connected to a single high performance, deterministic, low latency fabric. This eliminates hops between servers and opens up the platform to support intense E/W traffic within the servers that collaborate to deliver services. Contrast this to traditional architectures that put layers of switching between servers with in-chassis blade switching modules. The performance gains were noticed and communicated by Xerox customers to Tom, and that is the end result that really matters
Form factor agnostic design. In UCS a server is a server is a server regardless of the shape of the box. The Xerox MPS cloud leverages blade and rack servers as and where they make sense and the architects and administrators can manage them all in one abstracted pool of resources. No other platform so fully eliminates the concerns of what shape the sheet metal is.
A unified control plane exposed via XML API. The MPS cloud is orchestrated with vCloud director. The deep integration between UCS Manager and cloud platform SW enables automated discovery and configuration of new compute resources as they're added to the system. This creates the true elasticity and automation that a cloud of the magnitude of Xerox MPS demands. Programmable pools of abstracted computing and network elements is what separates a robust cloud from one built on a brittle, manual infrastructure foundation.
UCS Manager Service profiles: Simplification of server image types and elimination of configuration drift as applications move from development through test, staging and deployment was a big win for the Xerox IT team. Having a infrastructure that can be reliably and accurately provisioned and maintained, both in the primary and remote DR sites is another area Tom cited in our conversations.
UCS Central: this is recently released technology that allows customers like Xerox to manage multiple UCS domains across the data center and across geographies.
I'm sure i'm forgetting something but I'll go with 5 unique attributes for now.
This Thursday, the Xerox team is joining us for a dynamite webcast we've pulled together to talk about UCS and laying the right foundations for cloud. James Staten of Forrester, who is THE MAN on cloud, helps us kick it off and we also have architects from FICO joining to talk about their private cloud design. If you're in the business of looking at infrastructure strategy for cloud computing this is one you don't want to miss.
Back in March we announced the third generation of UCS, with significant expansions to the I/O and systems management capabilities of the platform as well as a new lineup of servers. This month we’re continuing to expand the UCS server lineup with the addition of four new models. The latest batch of M3 systems are comprised of three Intel Xeon “EN” class machines (E5-2400 series processors) as well as a four socket “EP” (E5-2600 series) blade server. Specifically: the UCS B22 and B420 M3 blades and the C22 and C24 M3 rack servers. These new servers round out the UCS portfolio with an even stronger set of products optimized for scale-out and light general-purpose computing as well as a new price/performance 4S category in the mid-range.
If you prefer watching than reading , here is a nice conversation between Intel Boyd Davis , VP & GM, Data Center Infrastructure group, Cisco Jim McHugh, VP UCS Marketing, and Scott Ciccone, Sr. Product Marketing Manager, highlighting the key benefits of these new models.
To figure out how these fit in, let’s step back and consider the broader evolution of server technology in play here:
1) Cisco has made server I/O more powerful and much simpler.
One of the key differentiators of UCS is the way in which high-capacity server network access has been aggregated through Cisco Virtual Interface Cards and infused with built-in high performance virtual networking capabilities. In “pre-UCS” server system architectures, one of the main design considerations was the type and quantity of physical network adapters required. Networking, combined with computing sockets/cores/frequency/cache, system memory, and local disk are historically the primary resources considered in the balancing act of cost, physical space and power consumption, all of which are manifested in the various permutations of server designs required to cover the myriad of workloads most efficiently. Think of these as your four server subsystem food groups. Architecture purists will remind us that everything outside the processors and their cache falls into the category of “I/O” but let’s not get pedantic because that will mess up my food group analogy. In UCS, I/O is effectively taken off the table as a design worry because every server gets its full USRDA of networking through the VIC: helping portions of bandwidth, rich with Fabric Extender technology vitamins that yield hundreds of Ethernet and FC adapters through one physical device. Gone are the days of hemming and hawing over how many mezz card slots your blade has or how many cards you’re going to need to feed that hungry stack of VM’s on your rack server. This simplification changes things for the better because it takes a lot of complication out of the equation.