IBM Virtualization Empowers a Dynamic Infrastructure

Competing successfully in today's difficult business climate requires a new approach to IT—one that accomplishes more while using less. Organizations must reduce costs, increase service levels, and proactively mitigate risks.

Virtualization is a major step forward—a keystone approach capable of empowering each of those goals. This is because, given a virtualized infrastructure, essential resources such as processing power, storage and network bandwidth need no longer be tied to specific hardware.

Instead, they can be allocated when, and where, they are needed in real time, to best address changing business conditions or strategies such as unexpected workload spikes, system failover, new service creation, service-oriented architecture (SOA) and many others. A virtualized infrastructure positions organizations better for change of all kinds, both planned and unplanned—and in today's business climate, change is the only constant.

Virtualization, then, is a central element needed to make an infrastructure more dynamic. And the more fully organizations virtualize, the more benefits—both IT and business benefits—are likely to accrue.

From the IT side, costs will typically fall, thanks to reduced complexity, superior resource utilization, recaptured floor space and lower energy requirements; service levels will climb; and systems and services will enjoy higher uptime and better performance.

From the business side, virtualization allows organizations to pursue new strategies in a simpler, faster, and more cost-efficient fashion. The success rate of those strategies can also be quantified more easily, and adjustments to optimize can be more straightforwardly executed. One of the key elements of the dynamic infrastructure—converged management of business and IT—is supported through virtualization in this way. Virtualization helps abstract out the technical details of the IT infrastructure, replacing them with a more holistic perspective on the business outcome. As a result, taking direct action to improve that outcome becomes much simpler.

Many forms of virtualization can be combined to render the infrastructure more dynamic
Exactly how each organization should pursue virtualization will, however, differ from case to case, depending on the specific assets, strengths and weaknesses, requirements and goals that apply.

The phrase "dynamic infrastructure" will never mean exactly the same thing twice; each organization must consider the different forms of virtualization and how they might help render its infrastructure more dynamic:

• Server virtualization: This form of virtualization allows a single physical host to run multiple virtual servers simultaneously. Because these virtual servers can be created at will on different hosts, and even moved from one host to another, the overall effect is to assign processing power to services as more or less of it is required in real time, rather than limiting processing power to a specific computer (where, in typical cases, it will not be utilized nearly as well as it might be). By leveraging fewer hosts for more processing power, server virtualization increases service performance and availability while reducing energy costs and generated heat. A second major benefit: valuable floor space in crowded data centers is freed.
• Storage virtualization: In the past, storage was treated as a static resource tied to specific systems. Through virtual storage, however, storage becomes a fluid resource that can be assigned when and where it is needed, anywhere in the infrastructure, in real time. If a given system has excessive storage available, it can be doled out to a service running on another system on an as-needed basis. This approach drives up service performance and availability, because it minimizes the odds that any given service will fail due to insufficient storage.
• Client virtualization: For many organizations, a major IT challenge lies in desktop sprawl—thousands of desktops that require all-too-frequent manual attention from IT team, particularly in the case of major rollouts such as a new operating system. Client virtualization solves this problem through thin clients, each essentially the same from desktop to desktop, which automatically access the necessary applications and data across the network. Management of those resources is now centralized, and user relocations are dramatically simplified because users can easily access their environments from any thin client in the organization.
• Application virtualization: Java applications running on dedicated servers may not always be able to respond ideally to unexpected spikes, and the outcome can be diminished performance or even service outage. A superior result will come from a virtualized application infrastructure that supports application workload reassignment across multiple hosts in a cluster, keeping applications up and running well even under the most demanding conditions.
• Network virtualization: There are several forms of this type of virtualization. By virtualizing network nodes, for instance, network capacity can be increased in proportion to the business need without necessitating more hardware. Multiple physical network switches can be consolidated into one virtual switch, reducing network complexity and costs. Virtual private networks deliver secure access to the organizational infrastructure from outside its walls, wherever employees happen to be.

Cloud computing: A powerful evolutionary step forward
Exactly how an organization goes about combining these forms of virtualization to achieve a more dynamic infrastructure will depend on its specific needs and strategies. However, one compelling option lies in creating a cloud environment, which combines several forms of virtualization for an exceptionally simplified, optimized result.

Via a cloud, entirely new classes of IT services can be created automatically, based on a Web request coming from a user; subsequently, the cloud does the work of creating and provisioning virtual servers, assigning them resources such as storage as they are needed, securing and managing services while they remain in use, and freeing their resources when the services reach end-of-life.

Because they allow services to be managed automatically at every stage in their lifecycles, requiring little to no attention from the IT team, clouds represent a major evolutionary step forward in the history of virtualization—fostering collaboration both inside and outside the organization, and generating enormous business value, while reducing costs and complexity.

IBM offers the industry's most complete range of virtualization solutions and services
As a leader in virtualization with more than 40 years of deep experience and proven expertise, IBM offers the most comprehensive portfolio of virtualization solutions and services available today.

• Virtual server management: IBM Systems Director delivers centralized management of everyday functions throughout a virtualized infrastructure, including server health, resource allocation, and others, for virtualized environments such as Xeon and VMware among others. It also integrates with the best-in-class IBM Tivoli service management portfolio to deliver more specific features and functionality as required.
• Virtual storage management: IBM TotalStorage Productivity Center centralizes and simplifies every aspect of managing a complex, heterogeneous storage infrastructure to increase both the uptime and performance of storage-dependent services. Ideally, it is paired with the IBM SAN Volume Controller, which pools storage across the infrastructure, transforming it into a fluid resource available on demand to any requesting service.
• Virtual clients: IBM can work with your organization to develop a customized thin-client solution capable of fully leveraging leading offerings from vendors such as Citrix and VMware, thus significantly reducing desktop sprawl and all associated management complexities and costs.
• Virtual application infrastructure: IBM WebSphere Virtual Enterprise drives the availability and performance of key Java applications by shifting workloads dynamically across clusters to correspond with emerging spikes in demand, thus driving up application infrastructure flexibility and scalability as well.
• Virtual networks: The IBM Virtual I/O Server shares physical I/O resources across AIX and Linux partitions on servers, improving network bandwidth for them as required in real time. Furthermore, the IBM BladeCenter Open Fabric Manager simplifies I/O management for BladeCenter chasses, also supporting instant network availability for new blades as they are added.
• Physical servers and storage: IBM offers a complete range of best-in-class physical hosts designed to drive server virtualization through extraordinary performance and availability, from the IBM System z mainframes to IBM Power Systems minicomputers to IBM System x servers for best-in-class x86 environments. And IBM System Storage offerings are extremely well suited to even the most demanding SAN and NAS infrastructures.

• Services: A full suite of IBM services are available to help organizations pursue virtualization, in whatever manner they require, from initial consultation and assessment through implementation and deployment to subsequent management and optimization—even including out-tasked, fully-managed services absolving the organization of direct responsibility.

IBM, in fact, is the only solution provider capable of delivering the complete range of technologies and services organizations require to leverage virtualization for the best business outcome—whatever their current circumstances, requirements, or goals.