New Fabrics for Consolidated Environments

Share it on Twitter  
Share it on Facebook  
Share it on Linked in  

Server virtualization and consolidation are great for the bottom line, but let's face it -- they wreak havoc on existing data center network infrastructures.


So it shouldn't be any surprise that vendors are falling over themselves to deliver integrated fabric and backbone solutions designed to accommodate the increased I/O pressure that comes from rapid server expansion, virtual or otherwise.


Brocade took a major step forward this week with the introduction of the DCX Backbone line, uniting the 48000 Director with 8 Gbps Fibre Channel connectivity on up to 896 ports to deliver a whopping 12 Tbps of throughput. The setup is also outfitted with the company's Adaptive Networking system that dynamically shares resources and adjusts bandwidth allocation to suit changing needs.


The DCX Backbone is the first fruit of Brocade's Data Center Fabric (DCF) architecture announced last summer. The idea is to combine storage networking and server clustering on a policy-based, application-driven framework that can leverage existing hardware and software resources to meet the increasing data demands of the modern enterprise. Expect to see new DCF-oriented HBA and File Area Network (FAN) systems later this year.


Emulex is also tapping into 8 Gbps Fibre Channel for its Virtual HBA line as a means to boost I/O capacity to suit heavily consolidated environments. The company combines its LightPulse Virtual HBAs with 10 Gbps Fibre Channel over Ethernet (FCoE) converged network adapters (CNAs) to boost connectivity between servers. The systems use N-Port ID Virtualization (NPIV) and Service Level Interface (SLI) to enhance migration between virtual machines using various protocols and transfer rates.


And for those of you who are not quite ready to launch into entirely new data center fabrics, new smaller-scale switching solutions are being designed with consolidation in mind. Blade Network Technologies' 10 GbE blade switches, for example, were recently cited by Frost & Sullivan for their ability to enhance server performance following virtualization.


Increased demand for improved network I/O is further proof that the data center is not simply a collection of discrete components -- it's an organic creation in which changes in one area affect many others. And one of the best ways to leverage the investment in virtualization (and the resulting savings in hardware acquisition budgets) is to ensure that the network can handle the increased data load.