Look before you leap: Understanding the converged infrastructure model

By Christopher Papa

The speed of business today is giving rise to a transformation of the data center. Traditional IT infrastructure silos create bottlenecks that cripple quality of service and increase operational costs due to slower workload deployment and more room for error.

Simply put, it can take too long to design, procure, deploy and integrate new IT infrastructure, and there are too many tools needed to manage it.

The pressure for change has never been greater. Increasingly, CIOs are looking to converged infrastructure solutions as a key to rapidly deliver IT services, maximize data center efficiency and strengthen IT service quality. In fact, Gartner estimates, “By 2015, one-third of all servers will ship as managed resources integrated in a converged infrastructure.”¹

Convergence, which combines compute, storage, networking and infrastructure management into an integrated system, is a strategy used by organizations to accelerate IT service deployment, increase efficiency and strengthen IT service quality by making the data center simpler, more flexible and more cost-effective.

These objectives are enabled by breaking down rigid, complex IT silos within the data center and transforming infrastructure into dynamic pools of server, storage and networking resources that can be shared by multiple applications and managed collectively using automated, policy-driven processes.

However, the transition from the traditional silo-based IT structure is something that will not happen overnight – especially as IT teams attempt to keep up with the evolving needs of business.

And, it’s not always easy to navigate through the various market solutions and impacts to an IT organization. It’s important for CIOs to understand various approaches and models to the converged infrastructure as well as tools and support that will be required as IT roles evolve.

Making the switch: Key considerations and avoiding pitfalls

There are multiple considerations CIOs and IT managers should be aware of when evaluating converged infrastructures:

  • Rule #1: Plan your converged infrastructure based on key workloads

When tearing down silos, IT must ensure the new converged system is designed to best address the needs of their business’s key workloads now and in the future – workloads such as analytical, unified communications, desktop virtualization and private cloud.

Forrester Research underscores this point in its recent report, “Optimize IT Infrastructure Around Key Workloads².” In the report, Forrester states, “The inside-out, technology siloed approach to IT infrastructure needs to stop. Instead, IT infrastructure needs to become workload-centric. In other words, design the server, storage, and networks in your data center on what matters most (your workloads), and not the other way around.”

Choosing a converged infrastructure platform that is pre-optimized for a broad range of workloads not only gives IT more agility to get systems up and running, but also lifts financial burdens with better total cost of ownership and lower operating expenses.

  • Converged infrastructure solutions should go beyond the hardware

It is critical to evaluate a converged infrastructure solution based on more than hardware alone. Software and services are essential elements that provide users with an open, intuitive, automated and end-to-end foundation that ensures IT can quickly harness the benefits of the solution. A single source of support for example can exponentially speed deployment and maintenance.

  • Deployment model options let customers choose their starting point and speed transition

It’s important to have the power and flexibility to choose what infrastructure best aligns with the needs of the organization. Some vendors offer deployment model options to help with the process, such as:

    • Do-It-Yourself – For administrators that prefer full choice and flexibility over their IT infrastructure, these “a la carte” solutions are offered with validated components that are interoperable with one another defined in the form of a support matrix.
    • Pre-Engineered Reference Architectures – Application-validated offerings help customers consistently build highly available converged infrastructure solutions for their critical IT workloads.
    • Pre-Integrated System – These solutions are designed for IT shops that want to take the guesswork out of building out a converged data center so they can deliver results to the business the fastest way possible. They are pre-assembled and optimized to be quickly operationalized.
  • Management software is a critical element

As IT marches towards convergence, using independent, standalone management software for every infrastructure sub-domain is slow and inefficient. This approach ultimately leads to missed business deadlines because of slow deployment processes and stalled IT initiatives.

A solid converged infrastructure solution should have unified system management software that collapses multiple management consoles, simplifies infrastructure configuration, and drives automation and consistency. This will enable IT to reduce time and steps to provision new workloads; efficiently allocate and manage server, storage and network resources; and quickly migrate or scale workloads as business needs change.

  • Look for an open and standards-based approach

Not all vendors take the same approach to offering holistic converged infrastructure, and CIOs should evaluate how the pros and cons of each will impact their strategy. For example, while some converged infrastructure solutions come from a single vendor, others are actually grouped together by multiple vendors.

Many times, when multiple vendors are involved, organizations must rely on consultants to help determine whether they can actually mix and match hardware and software within a system. The process can significantly slow down implementation and system upgrades and increase costs due to consultant fees.

On the other hand, companies are hesitant to adopt single vendor solutions because they want to avoid issues associated with vendor lock-in or they want to leverage some of their existing infrastructure in their converged solution. When choosing any platform, IT managers should ensure the solution is backed an open, standards-based approach to achieve optimal flexibility and interoperability.

Converged infrastructure promises to enable IT to better meet business goals, ensuring operational agility, efficiency and quality of IT service delivery well into the future.

While not necessarily the answer for every organization, many companies are already reaping the benefits of convergence and choosing platforms that are optimized for their most important workloads and applications.

Top solutions providers recognize the need for workload-centric platforms, and organizations can look forward to even more solutions to help them achieve the benefits of converged infrastructure.

If your organization is evaluating converged systems, remember that one model won’t fit every need; solutions should align with your strategy, not the other way around, and convergence impacts the entire data center – infrastructure, management, applications and service management – not just the hardware.

The author is the country manager of Dell Philippines

¹Is the Concept of the ‘Server’ Obsolete, or in Need of Redefining? 29 March 2012
² Optimize IT Infrastructure Around Key Workloads, Forrester Research, Inc., September 7, 2012

Comment on this post