Data Center

On-Demand Data Center Validation: The Beginning

by mschiff on ‎09-11-2013 08:37 PM - last edited on ‎10-28-2013 01:24 PM by bcm1 (3,923 Views)

We find ourselves at a tipping point for a new data center network. Demands on the modern data center have completely outpaced innovation on the network. Just like with anything nowadays, (movies on Netflix, music on Spotify, rides on Uber,burgers from McDonalds) it’s all about on-demand. It is now time for the ability to flexibly deploy data center capacity – compute, networking, storage and services - in real-time, whenever and wherever you need it.


In April of this year, Brocade unveiled its vision for the On-Demand Data Center to meet this need. This blueprint addresses the technology and business imperatives we hear most frequently from data center operators: virtualization, multi-tenancy, return on investments, and cost savings.


Today we embark on a series of blogs that will show how this is not just a vision.  You can achieve results from your data center network today. Over the course of this series I will take you into our laboratory and show you how you can:

  • Reduce network traffic congestion by up to 40%
  • Achieve 100% link utilization with no packet drops
  • Create and move VM's on-demand with no network downtime or reconfiguration
  • Realize 100% isolation of both tenants and application traffic
  • Reduce network provisioning time by 90%
  • Deliver new services 80% faster
  • Get 100% visibility into your network


Emerging and exciting technologies like SDN, NFV, and Ethernet fabrics will be covered, but only in ways that can be deployed today.  This is not a 12-18 month roadmap.  These are validated solutions through test cases, which comprise the On-Demand Data Center shown in the network diagram below.




In the next blog, I will focus on the effects that virtualization has had on the data center network.  Then, through test cases, you can see how to optimize east-west traffic flows, increase automation and elasticity of workloads, and reduce congestion caused by virtualization through efficient network designs.