Turning the Network into a Fabric

Arthur Cole
Slide Show

Seven Best Practices for Virtualization

Virtualization is taking IT to new horizons from which whole new sets of opportunities are coming into view.

Do you have a fabric yet?

It seems that the only way to take full advantage of virtualization, the cloud and just about anything else that comes along is to convert today's stodgy, old enterprise network into a sleek, new fabric. But exactly how does one go about this? And how will you know you have a fabric when all is said and done?

Clearly, the need for network enhancement is prevalent in the virtual age. Once data and applications are delinked from underlying hardware, the movement of that data from place to place becomes the key factor in advanced architectures.

The ideal fabric will consist of a high-density, non-blocking 40 GbE infrastructure that provides things like multipath support and standards-based provisioning, preferably on Layer 2 to better serve the cloud, according to Extreme Network's Shehzad Merchant. Along the way, you'll need to ensure that the fabric provides active-active data pathways with minimal hops to ensure bandwidth efficiency and fast failover capabilities. In the end, you're striving to replace current network topologies like Spanning Tree with a much more flexible mesh design.

Recent weeks have seen a steady stream of new and enhanced fabrics, each one claiming to push the envelope when it comes to data flexibility. Enterasys jumped into the ring with the new OneFabric designed to cut network overhead some 90 percent even as it scales up to meet new virtual and cloud requirements. The system offers centralized control for improved workflow management and enhanced application delivery for VMware, Citrix and Microsoft virtual infrastructures. It also has the ability to extend into mobile environments.

Of course, where would Cisco be without a fabric solution? The company recently enhanced its Unified Fabric portfolio with a new line of Nexus 7000 switches and support for the FabricPath architecture on the Nexus 5500 switch, moves which extend support to more than 12,000 10 GbE ports. The package includes new sub-microsecond latency technology for the Nexus 3000 for 40G L2 and L3 deployments. The company has also added IPv6 support to the Application Control Engine (ACE), providing additional addresses and enhanced routing for Internet-capable devices.

Meanwhile, Brocade is out with its Virtual Cluster Switching (VCS) platform on the VDX switch line. Designed to provide a self-healing, self-configuring network environment, the system enables masterless automation for either full or partial mesh configurations. At the same time, it increases I/O throughput at the access point and then flattens both the access of aggregation layers to reduce network overhead. The company says this is more efficient and less costly than basing the fabric on the core switch.

In the very near future, enterprises will no longer have to worry about provisioning adequate resources to handle data loads. Processing and storage will always be available somewhere to handle the peak.

That means organizations will have to build networks capable of handling the fluid nature of new virtual and cloud infrastructures. Fabric technology offers the most advanced architecture to accomplish this, even if it means reworking one of the most fundamental aspects of the data environment.

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.


Add Comment      Leave a comment on this blog post

Post a comment





(Maximum characters: 1200). You have 1200 characters left.




Subscribe Daily Edge Newsletters

Sign up now and get the best business technology insights direct to your inbox.

Subscribe Daily Edge Newsletters

Sign up now and get the best business technology insights direct to your inbox.