Imagine a job that requires managing complex, multi-tiered IT environments every day-often on a global basis. There is constant pressure to maximize resource usage while simultaneously reducing management costs and maintaining a highly-available, secure environment, all within strict budget constraints. Sound easy? These are just a few of the challenges data center managers and network administrators face every day.

An array of products has emerged to enable the virtualization of the data center at the server and storage device level, and fortunately, the value, characteristics, and benefits of virtualization can also be utilized in the physical infrastructure layer. A virtualized infrastructure layer not only enables the endpoint virtualization applications, but also simplifies and automates IT tasks to help reduce management expenses and overall total cost of ownership.

Figure 1. The above illustration shows how the infrastructure layer can be ‘virtualized’ such that all changes, migrations and monitoring are automated, scalable, and integrated while the overall network management is simplified, more robust, and more secure.


To fully adopt a virtualized data center, the physical infrastructure layer (switching, patch panels, monitoring equipment, and cabling) cannot be ignored. Typically, data center networks comprise some combination of SAN, LAN, WAN, and MAN components. However, regardless of the mix of IT equipment and network protocols and interfaces, data center networks require a flexible and adaptable infrastructure layer to connect physical and virtual devices together.

A virtual infrastructure layer (VIL) allows for automated and secure equipment additions, moves, or changes, thus enabling seamless network reconfiguration and growth. Adapting to the dynamic needs of a virtualized data center, a VIL provides a more secure environment through its “hands-off” network management approach. It also provides resiliency, with monitoring and diagnostic tools that assist in managing the network.

Virtual Infrastructure

The essential characteristics of the virtual infrastructure layer are: automation, security, performance profiling, monitoring, network resiliency, and physical media transparency.  A virtual infrastructure switch provides an integrated solution that addresses each of these components.

Traditionally, physical network connectivity is achieved through passive, labor-intensive, and error-prone manual processes that rely on physical patch panels. In many cases, physical cable lengths and complex, inaccessible cable routings make data center evolution nearly impossible. Years or mere months, sometimes, of cable accumulation often results in inefficiency and inability to grow or consolidate infrastructure.

The core of a VIL is a non-blocking, high-availability switch that can scale to several thousand ports. This automates the physical layer so that minor changes are reduced to seconds; major changes that once took days can be made in minutes - all without ever having to go on the data center floor. A VIL also provides the capability to quickly add or migrate IT equipment as needs evolve without significant downtime and risk. The data center is then able to scale easily and become future proof against unanticipated changes.

For example, if storage needs unexpectedly increase, new equipment can be installed, connected to the switch, and tested prior to being brought into the production network. The new equipment can be quickly merged into the network, and similarly, older IO devices can be transitioned out in such a way that disruption is minimized and migration is simplified. The ability to efficiently share resources and lower overall IT costs is another key value. Expensive data center equipment, purchased on a department or division basis, can be shared based on time-sliced needs between departments, divisions - and in some cases continents.

Eliminating the need for physical access to the infrastructure significantly reduces the possibility of malicious or unintentional security breaches. With a virtual infrastructure switch, all adds, moves, and changes to infrastructure are completed via software that remotely controls the digital connectivity in the virtual infrastructure switch. Unintended changes can be reversed very simply. In addition, the switch supports hard partitions or virtual fences to segregate and isolate network ports in a shared environment to minimize physical access to IT equipment and create a centrally managed, “hands-off,” ultra-secure environment.

Many tools attempt to balance network loading, but common issues still occur. Often, unbalanced network situations are left unchecked, resulting in IT equipment degrading to its breaking point or being underutilized. Complete visibility with such infrastructure is simply not available. A virtual infrastructure switch leverages its central location in the network to provide end-to-end visibility and statistical information, which administrators can easily interpret and act upon to balance network and resource utilization.

Typically, scalability issues occur when standalone diagnostic tools are connected in-line with problematic paths in the network. This equipment needs to be placed actively and disruptively into the data path, thus changing its characteristics (latency, bandwidth, etc.). The virtual infrastructure switch provides a far more integrated and scalable solution to monitoring and diagnostics. Administrators can route any path(s) within the infrastructure passively into a desired test device without ever moving a physical cable or changing its end-to-end circuit characteristics. Further, a virtual infrastructure switch provides increasing levels of diagnostic and tracing facilities that are integrated into the product and act quickly to expose and enable correction of network problems.

Diagnostic tools within the virtual infrastructure switch offer the capability to scan data paths and capture protocol traces automatically on error conditions or the ability to test network reaction to loss of signal by issuing a test command. The switch merges network performance monitoring and diagnostics with the automated infrastructure function. Management interface tools need to provide visibility into the virtual infrastructure including the supported networks providing alarms in multiple formats when issues arise. In addition to visible alarms, SNMP alerts and e-mail home capabilities are provided.

Figure 2. Virtualized infrastructure breakdown

It is imperative that enterprise-class IT equipment guarantee a fault tolerant architecture providing at least five nines (99.999 percent uptime) availability. This requires that no devices in the network have any single points of failure (redundant ac power, dc power, functional failover/fail-through, etc.). Often, what is forgotten in the data center design is that infrastructure failures are among the most common. Components including cables, patch panels, and optical interfaces are often the most frequent failures in the network resulting in points of vulnerability that weaken the network chain.

Resiliency must be built into the physical infrastructure layer to help ensure maximum availability, and administrators should be able to configure failover paths for critical data connections. Then, if an infrastructure component fails, operations will automatically roll over to the alternate and continue to function seamlessly.

In a fiber environment, marginal signals are a common cause of intermittent problems and can be costly to isolate. With each patch and bend in an optical cable, there is signal attenuation. With a VIL, signals are monitored and re-driven at each port, enabling a more resilient physical layer in the data center. The virtual infrastructure switch can also be leveraged to extend distance via long-haul, small-form pluggable (SFP) optics.

As the network grows, boundaries and physical interface mediums change. For example, a business unit may expand beyond the confines of a room, floor, or building requiring different network interfaces for connectivity. In the case of expanding infrastructure connectivity outside of a building, it may be necessary to expand from local copper cabling to campus metropolitan area network (MAN) distance optical devices.

VIL solutions create a media-agnostic environment by allowing connections through the infrastructure switch, seamlessly adapting diverse physical media. In order to implement a VIL that is flexible, each port on the infrastructure switch needs to allow for SFP devices that can be copper, short-haul optical, or long-haul optical for network interface connectivity.

Physical Space

With all of the beneficial characteristics of a VIL, it is important to keep in mind an overlying physical advantage in extending the spatial, cooling, and migration needs of the entire data center. Rat’s nests of cables under raised floors contribute to cooling as well as growth problems, and it is necessary to keep in mind that data centers are dynamic, not static.

Common Applications

There are numerous applications already reaping the benefits of VIL systems, including disaster recovery and business continuance needs, where recovery and backup times are significantly reduced. Remote data center infrastructure management is also a major benefactor, allowing IT professionals to “follow the sun” and keep systems running no matter where they are in the world. Key industries currently include government agencies, Fortune 100 retail companies, transportation firms, technology test labs, and more. The more informational transactions that a company makes the greater the need for the ability to wire the data center once and then walk away.


Today’s data centers are vulnerable to human error, storage growth rates, infrastructure and monitoring issues, and more. Companies need highly available, fault-tolerant, scalable, physical infrastructures to support their mission-critical LAN, SAN, MAN, and WAN network environments. As IT dollars become tighter and data center environments become more complex, a VIL simplifies and secures the underlying network cabling and its management. Quantifiable in many ways, the net benefits include significantly reduced change times, ease of migration, faster troubleshooting, and enhanced resource utilization, visibility, and security that results in lower overall network management and data center infrastructure costs.