IT Roadmap

Cloud is Not a Matter of If, But How Much and How Fast

During the last 18 months, Cisco has been conducting customer round tables and research to uncover the themes driving enterprises to transition their network storage and compute infrastructures to the cloud. One of Cisco's forecasts, based on this information, is that 15% of IT workloads will migrate to a cloud-based infrastructure by 2013.

In this video, Chris Osika (Senior Director of Cisco's Internet Business Solutions Group Sevices Provider Practice) shares 4 emerging themes in relation to the cloud:

In summary, the four themes are:

1. Sheer Impact. In the past, it could take up to 45 days to provision a server. Now, companies are reporting an hour or less with data center virtualization. That's a pretty dramatic change.


2. Organizational Change. Companies are addressing the need to re-align in a functional way to take advantage of of virtualization and cloud. When they do so, they are able to enact a fully holistic business transformation.


3. The Ability to Flex. Virtualization with cloud computing not only allows companies to flex compute and storage resources, but to flex the network to reap rewards of economy.


4. Virtual Desktops (VDI) Enter the Discussion. Whereas in the past, the discussion of VDI was limited, it's now included in conversations about the impact cloud computing can have on the overall enterprise.

Do these themes ring a belll with your company? What other themes do you see emerging  in relation to accelrating your company's presence in the cloud?

6 Levels of Opportunity with Virtualization

It's easy to get lost in the world of virtualization. After all, the word is increasingly being applied to all sorts of situations and applications.

Recognizing the problem, Daniel Kusnetzky of the 451 Group has created an accessible framework that sorts out the various layers of virtualization that deserve our consideration. Each level of the model represents a different opportunity to implement virtualization.

According to Kusnetzky, virtualization can be applied to address "performance, reliability/availability, scalability, consolidation, agility, a unified management domain or some other goal." As he explains, each level offers different benefits:

• Access Virtualization — hardware and software technology enabling just about any device to access just about any application. While the application can see the device, the device knows how to display it. Special purpose hardware may be necessary on each side of the network connection to enhance performance, enabling many users to share a single client system or allowing a single individual to see multiple displays.


• Application Virtualization — software technology enabling applications to run on many different operating systems and hardware platforms. This usually means that the application has been written to use an application framework. It also means that applications running on the same system that do not use this framework do not get the benefits of application virtualization.


• Processing Virtualization — hardware and software technology that hides physical hardware configuration from system services, operating systems or applications. This type of virtualization technology can make one system appear to be many or many systems appear to be a single computing resource to achieve goals ranging from raw performance, high levels of scalability, reliability/availability, agility or consolidation of multiple environments onto a single system.


• Storage Virtualization — hardware and software technology that hides where storage systems are and what type of device is actually storing applications and data. This technology also makes it possible for many systems to share the same storage devices without knowing that others are also accessing them.


• Network Virtualization — hardware and software technology that presents a view of the network that differs from the physical view. So, a personal computer may be allowed to only “see” systems it is allowed to access.


• Management of virtualized environments — software technology that makes it possible for multiple systems to be provisioned and managed as if they were a single computing resource.

While these technologies have been available in various forms for almost three decades, they are now widely available for industry-standard applications. Such availability is clearly contributing to the ongoing wave of virtualization that is transforming IT.

Virtualization to Grow to 1 in 4 Server Workloads by Year-End

Gartner is estimating that one out of every four server workloads will be virtualized by the end of this year. However, the research firm argues that enterprises should invest more in the approach if they want to maximize the return they are getting on their IT dollars. 

"More than 80 percent of enterprises now have a virtualization program or project, but only 25 percent of all server workloads will be in a virtual machine (VM) by year-end 2010," Gartner reports. Virtualization, the firm adds, is "the highest-impact issue challenging infrastructure and operations."

Virtualization was first introduced decades ago as a means of managing mainframe technology. More recently, it was adopted as a means of enhancing the usage of x86 servers, which comprise 90% of the server market overall.

Now, server virtualization is actively employed throughout the Fortune 500. IT executives consider it a huge win because they can drive server utilization rates from as low as 10% to as high as 70-80% — producing huge savings in the process.

However, some companies remain reluctant to plunge into server virtualization. They cite security concerns, performance issues and software licensing complications among the issues holding them back from embracing virtualization technology.

Despite these concerns, some research firms remain bullish on the technology's adoption. One firm, The InfoPro, reports that more than half of newly deployed servers are virtual machines and predicts this number will climb to 80% by 2012.

That said, the implication of Gartner's prediction — covering "only" one-quarter of server workloads — is that three-quarters of workloads on physical machines have not been virtualized.

This raises some questions about whether executives are investing their IT dollars as smartly as they might. In fact, Gartner argues that eight of every ten dollars that enterprises spend on IT can be considered "dead money" because it does not contribute directly to business value.  

“We say ‘dead money’ because, while it is keeping the lights on, it isn’t directly contributing to your business growth or enhancing your competitive advantage,” says Daryl Plummer, managing vice president and Gartner Fellow. “In today’s environment, any corporate function that doesn’t contribute to growth or competitiveness is ultimately expendable. Your placement of resources is more critical than ever to your ability to deliver the growth and competitive advantage that your CEO is expecting.”

Putting Business Strategy Before Cloud Strategy

The hype around cloud computing is enormous. But the danger for IT decision makers is that they will get swept up in the mania and drawn into making investments that aren't particularly consistent with their business objectives.

Given these risks, Randy Heffner, vice president and principal analyst at Forrester Research, is encouraging his clients to step back and take a few breaths. "Cloud is an important development in the landscape of computing options—to the point that most organizations will one day use cloud or cloud-like offerings—but there's no guarantee that cloud is right for your organization right now," he says.

While he recognizes that IT decision makers may be tempted to create their own "cloud strategies," he is discouraging it. As he sees it, enterprises can lose focus on what's important when they develop strategies around technology as opposed to business.

What's more, he sees "numerous tradeoffs" between cloud and traditional computing approaches. "Nearly every cloud solution has a functionally equivalent non-cloud alternative, so to maintain focus on your business, it is best to build your strategy around the business decisions to which each type of cloud offering is directed," he contends. "This approach fosters more level-headed consideration and comparison of cloud and non-cloud options, and it establishes a stronger foundation for a long-term evolution toward cloud and cloud-like options as they mature."

Heffner encourages IT decision makers to develop their business approaches around three major architecture strategies: computing infrastructure; application platform; and solution portfolio. 

"Rather than developing a siloed strategy for cloud, a business approach will integrate analysis of cloud and cloud-like options into each of these three strategies," he says. "Rather than asking architects to create a cloud strategy, CIOs should direct them to enhance existing strategies. This will include the development, within each strategy, of plans for evolving over the long-term as cloud offerings mature. This reframing away from cloud strategy is much more than semantics. By centering analysis and strategy on business problems and decisions," he explains, the IT decision maker:

• Frames the full range of business criteria by which to compare cloud and non-cloud options.
• Moves analysis away from theoretical definitions of what is and isn't "pure cloud," allowing more cogent analysis of cloud-like, hybrid, transitional, and traditional computing options.
• Allows clearer matching between the economic and risk profiles of a wide range of computing options and the business and financial dynamics of different business scenarios.

Heffner offers specific advice with regard to positioning cloud offerings in relation to concerns about infrastructure as a service, platform as a service and software as a service. But the bottom line is that IT must ensure the technology cart does not get in front of the business horse.

Improve Customer Interactions with UC Plus Virtualization

Customers demand easy access to information—through self-service Web portals as well as from customer service representatives. Customer-focused organizations can improve customer interactions dramatically by integrating communications systems with customer information.  

Data center virtualization, in combination with unified communications systems, improves customer interactions, provides customer service agents with real-time access to up-to-date customer data, offers intuitive self-service options, and protects availability and security of valuable customer information.

Data center virtualization is the unified “back-end” for a Unified Communications platform. Virtualized data centers offer ubiquitous, always on access to critical customer information. The seamless integration of applications and data services within a virtualized IT infrastructure enhances any customer service or call center operation.

Take a look at 4 Advantages:

• Always On: Virtualized application services assure that customer service applications are always available to self-service customers or customer service agents. There is no need for expensive redundant data centers or failover operations, since the virtualized infrastructure is inherently redundant.


• Up-to-Date: Virtualized data services assure that customer information is always up-to-date and in sync regardless of access methods (e.g., web or CRM) or application instance.  Customers and service agents enjoy real-time customer information from any access point.


• Ubiquitous response:  Whether a customer is signing in from New Delhi or New York, virtualization enables them to be routed to the most efficient resource to meet their information needs.  Call-center agents are also routed to the most efficient processor within the virtualized infrastructure assuring optimum performance and response times regardless of locale or current application/information workloads.


• Reduced Cost: The cost of providing customer information application services is dramatically reduced, thanks to a reduction in CapEx and OpEx  costs  associated with the procurement, appointment and ongoing management of a virtualized IT infrastructure.

When combined with a Unified Communications platform, data center virtualization empowers customer call centers or self-service customer applications to deliver the information needed quickly, reliably and in real time, wherever it’s required.

Virtualization Offers More than Scale

Explosive growth of data and processing requirements is driving the need for a new architectural approach to IT infrastructures. The approach of mapping single servers to specific applications simply cannot meet the increasing demands for resources and dynamic response to business needs. Data center virtualization is the solution.

Virtualized architectures offer scalable and cost effective processing platforms. However, that just scratches the surface. Beyond the obvious benefits of scale, virtualization offers:

• Agile response:  Dynamic businesses require quick response. Thanks to virtualization, IT teams can respond to new user and business demands for applications, services and more. Changes in applications, workflows, software revisions, desktop configurations and more are straightforward to deploy and immediately available across the entire infrastructure.


• Reduced capital and operating costs:  Virtual platforms deliver new levels of utilization and performance. Servers that were once 30% utilized reach 80% plus within a virtual infrastructure. With this increase in ROI, costs for power, space and capital are reduced. Management costs are also dramatically reduced, allowing IT staff to focus on innovation and next generation services instead of day-to-day system and resource management.


• Business Continuity:  Virtualized architectures offer affordable alternatives to traditional redundant data centers. Virtual resources act as backup resources across the infrastructure, enabling planned and unplanned downtime with minimal, if any, impact on service delivery.

Virtualization isn’t a new concept. Service providers offered virtualized access to outsourced applications and services nearly a decade ago. Today, a wide number of vendors offer virtualization software, off-the-shelf hardware components and networking solutions that enable an IT staff to integrate and create a virtualized data center.

But the benefits of virtualization are not all in place, yet. The movement to blade servers and storage appliances improves overall ROI while simplifying management. Reduction to costs is important – but there is more to come.

The ultimate benefits of virtualization will come from an automated, ‘lights out’ approach to delivering IT services. Early tools for automation are being deployed by a number of independent vendors. The rewards from virtualization will continue to increase as these tools come to market, are improved and refined, and as they reach ‘standard’ levels for interoperability.

Still, for the near-term, the best approach to data center virtualization is to work with partners who are experts in the design and deployment of virtualized infrastructures.

It’s clear that the time for virtualization is now. The opportunity to improve results from reductions to capital expense, ensuring business continuity and improved agility that enables IT to quickly meet business needs are definitely within reach.

Answer 5 Questions to Ensure Your Network Supports Virtualization

Implementing data center virtualization and virtual server networks isn't as simple as install and done. One of the pivotal components is your network infrastructure. Although virtualization offers significant benefits when it comes to server and desktop deployment, management and utilization, the network requires some updates in business networks to be most effective at supporting the increased demands of virtualization.

Network optimization is critical in virtual environments.  If networks are not properly architected, with the required tools for management, virtual infrastructures can result in less efficiency and more cost. Something every IT department is tasked to avoid.

Following are 5 Questions to answer when planning network support for a virtual infrastructure:

• How will the change affect staffing requirements, and what will be the cost of added complexity, such as increased downtime to accomplish the migration?


• How will you manage your virtual network? While there might be fewer physical switches and routers, a virtualization strategy often means more logical, virtual devices that also have to be managed. Many traditional networking tools may not spot problems unless specifically created for a virtual environment.


• How will you simplify network management? Does your tool allowing IT professionals to interact with a virtual network in much the same way that they would interact with a physical network, such a product should make supporting server virtualization similar to what they already do.


• Can you scale to manage more and diverse devices? Moving to data center virtualization almost always decreases the number of physical servers, but the number of virtual storage servers, for example, can more than double. [International Data Corporation]


• How will you automate network tasks? Virtualization doesn't affect what network managers do, per se. But it does increase how often management tasks must be performed, while requiring more advance planning for capacity and network load balancing. Automation of these tasks is a key to improved efficiency.

Most experts agree that network planning is one of the most critical aspects of any Data Center Virtualization project.  That’s why having a partner that understands how business networks and virtualization work together is critical to success.

Don't Forget About Your Network

The uptake in data center and server virtualization requires that IT rethink their data center networks.  Virtualization requires a new approach  - and perhaps a new vendor choice.

Network optimization is a key issue when it comes to virtualization.  Why? Because virtualization increases server utilization, which in turn increases traffic - creating I/O congestion from the edge devices all the way into the data center.  Within the data center, network demand is geometrically increased as virtualized workloads are distributed across multiple servers. 

Vendors are targeting the network needs of virtualized data centers with a variety of solutions.  These divergent approaches to delivering next generation data center networks can be confusing – and proprietary.  Consequently, choosing the right virtual vendor can be more difficult than ever. 

It’s unclear today which vendor will dominate the virtualized data center network market. What is clear is that IT needs to be smart when selecting next generation technologies:

1. Integrated is better than standalone. Virtualization requires tight integration and scalability between the network and the virtual server architecture.  Buying separate products and knitting them together may be lower cost up front – but will most likely by higher cost in the longer term thanks to the ongoing management overhead to continually tune the products to work together.  Not to mention the finger pointing.  Solutions that are designed to work together from the network through the virtual server architecture will improve initial deployment and ongoing management (and productivity).


2. Avoid vendor lock-in and costly upgrades. As with previous technology advances, proprietary approaches will rule for the first phase of solutions. Each vendor will offer their approach – and over time, will converge to offer open solutions (or integration APIs) to achieve interoperability.  If history repeats itself – market leaders will set the open standards – meaning that buying from an unknown vendor may leave you stranded in your own locked-in world.


3. Double check that roadmap. Ask to see any vendor’s roadmap – as well as for references on their current deployments.  Don’t buy anything without customer evidence of its value.   Ask your network experts to review comprehensive technology roadmaps with each vendor to be sure that vendor is heading in the direction you want to go.

Most importantly – choose a vendor who understands your requirements, your business and can show you successful deployments in customers who mirror your own data center and infrastructure requirements. 

IP Telephony is More Reliable Than You Think

Just how dependable are Cisco voice over IP solutions?  This uncertainty is why many people get concerned when they find out they will be using a phone that is attached to the data network. 

These concerns are rooted in several sources.

• Historic network performance – computer networks in general have the persona of being somewhat reliable.  Many consider the performance of an Internet connection as what they could expect of a new IP phone system. It’s the “somewhat reliable” that causes concern.


• Perceived network performance/PC performance – websites, applications or the PC appear to be sluggish at times.  While this could be due to any number of factors, to the typical end user, this is the network.


• Reliable phone systems – there is no doubt the telecommunications industry has created traditional phone systems that are extremely reliable. This minimizes the need to switch because they don’t understand the benefits beyond dial tone.

There is an industry joke that if a network engineer maintains a data network with 95% uptime, he is a hero.  If a telecommunications engineer maintains a telephone network with 95% uptime, he is fired! 

However, given the advancements that Cisco has made in this technology space, systems can now be designed and configured to provide better reliability than traditional phone systems.  The very basic concept of IP telephony allows for tremendous flexibility, scalability, and more importantly, reliability. 

Traditional phone systems have all the functionality built into the PBXs – configuration, call processing, call routing, voicemail, PSTN connectivity, and handset connectivity – to name a few.  IP telephony breaks these functions out into separate devices that can be dispersed throughout the network, as well as duplicated to provide redundancy.  Now you can have a voice gateway with PSTN connection in 2 offices and provide backup PSTN connectivity to each site or have call processors in the 2 locations to provide backup call processing and call routing to ensure that you always have the ability to send and receive phone calls.

Aside from utilizing redundant equipment, most recent voice server applications use the concept of active and in-active partitions.  This now allows for more efficient upgrading of the application but also a fall-back plan if the new upgrade doesn't perform as expected – all with minimal downtime (recall that we are using redundant equipment as well!)

Cisco’s latest development has been to move the voice application servers to a virtual environment.  This not only reduces operating cost, but also provides a robust platform for streamlining the deployment of voice application servers making the switch from a traditional telephone system easier than ever..

So what are you waiting for?  Find out how your company can benefit from Cisco Unified Communications today!

Today's post written by: Nathan Richie, boice.net Network Consultant

Tackle Key Challenges for Data Center Virtualization

As many enterprise IT departments begin the transition to virtualized data infrastructures, a few main challenges are leading the pack. 

A recent study of top IT executives ranked the top two challenges as:

1. C-level executives protecting tight budgets and
2. Software vendors that don't support their applications on virtualized servers.

It’s interesting to note that these issues ranked as top challenges whether transforming a legacy data center or building a new one.

The survey results also supported already well-known ideas that virtualized systems invariably save time, power and money once they're deployed.

"Virtualization has become a significant asset for IT, reducing time and cost of deploying new servers. However, the greatest challenge has been resistance from software vendors [reluctant] to support their applications on virtualized servers," one respondent said.

"I certainly believe that IT virtualization makes it easy to manage databases more efficiently and effectively," wrote another respondent. "However, we need to convince our top management—the decision-making point—with return on investment, the cost savings in hardware and reduced administration costs."

Finding good people with the right skills for the new generation of systems is another recurring problem, according to the report. Organizational challenges increasingly arise as participants plan to implement various virtualization strategies, requiring a variety of skill sets.

Three out of four respondents say tight budgets are the greatest issue their IT operation currently faces. Even when the economy improves and budgets become less constrained, there will still be challenges facing IT managers.  

Here are the Key Steps to Focus on When Planning Your Transition:

Take a hard look at your applications. If your mission critical applications aren’t virtual-ready, then build a plan to test virtual resources with applications that are virtualized. In the meantime, meet with your application vendors and develop a plan to configure their application for your needs. If you have custom applications, start developing the code you need to optimize them within a virtual deployment.

Create a compelling plan, and sell it. Great virtualization projects need fuel—and they need a plan. So focus on planning early in your process.  

Create a plan that phases your deployment – starting with an early test, then a larger Proof of Concept (PoC), then a small production deployment followed by expanded roll out.

Set measures of success – show what the ROI will be for each phase and then demonstrate it. Tie budgetary support to meeting these measures so there’s a plan supported with a financial commitment.

Begin recruiting now – or train your teams. As with any architecture or technology, experience is vital to success. With a new architecture – you need training and expertise to get up to speed. So find the experts who can train your teams and invest up front.  You’ll be more successful in the long run.

By using the experiences of the leaders as examples, you can more easily move forward with your own virtualization deployments. Implementing data center virtualization doesn’t have to be a difficult transition. With the proper planning, budget and supporting expertise you’ll master “virtual” in no time.