Richard Napolitano

Subscribe to Richard Napolitano: eMailAlertsEmail Alerts
Get Richard Napolitano: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn

Blog Feed Post

The Next Generation Agile Data Center and the Birth of A Dynamic Network

At Plexxi we’re building a simply better network for the next era of IT to deliver agile data centers, enable scale-out applications and support distributed Cloud deployments.  In my prior blog, I discussed why the decades-old practice of pre-architecting, designing and implementing static network infrastructures wouldn’t support the dynamic needs of organizations moving forward. In this installment of my blog, I will review a case study from a large enterprise deploying an agile data center to meet their needs for the next generation.

I continue to spend a lot of time on the road, and I enjoy meeting with customers to make sure I stay on top of next generation data center networking requirements.  I recently visited a large enterprise that was experiencing scalability, management and performance problems with their existing data center network.  As the number of virtualized applications and corresponding virtual machines (VMs) grew in their data center, agility was actually decreasing rather than improving.  The data center was harder to manage due to application mobility and a lack of unified visibility across their virtual and physical environments.  In addition, their oversubscribed switches were experiencing capacity and buffering problems, caused in large part by the bursty and unpredictable traffic patterns of a virtualized data center.  While well designed for legacy north/south traffic patterns resulting from static applications, their multi-tier network architecture was not meeting the needs of their rapidly growing east/west traffic resulting from VM to VM traffic.  They had an upcoming project requiring the installation of over 30,000 additional VMs to support a virtual desktop infrastructure (VDI) deployment and they realized the existing network would not be able to support it.  They needed to make a change.

As they set out to research their options, they created a list of desired features/functions, outcomes and evaluation criteria.

  1. Ease of use
  2. Ease of deployment
  3. Ease of scalability
  4. Centralize control (SDN or other)
  5. Utilize dynamic pool of network capacity (similar to compute and storage)
  6. Reduce hardware rack space
  7. Reduce power consumption
  8. Reduce cooling requirements
  9. Deliver high performance – no oversubscription (1:1)
  10. Support for OpenStack Integration with vSphere
  11. Support for Virtual SAN

They evaluated Cisco, Arista and Plexxi’s technology solutions, and based on their success criteria and desired outcome, chose to deploy a next generation network system consisting of Plexxi Control, Plexxi Switch 2 and Plexxi Connect.

They decided to move forward with a brownfield deployment strategy, leveraging a cap-and-grow approach rather than rip-and-replace.  This approach would provide a less disruptive transition.  They migrated their virtualized applications and virtualized storage onto the new Plexxi network, and redundantly connected the Plexxi Switch 2s into the existing network; which was used to support their legacy static applications.

The Plexxi Switch 2 is two Rack Units high (2RU), taking up significantly less rack space and using much less power than traditional multi-tier networks comprised of large, expensive and power hungry chassis.  The Switch 2 provides redundant connectivity for the data center servers and storage appliances while creating a single tier, highly meshed, and high performance network.  Network scalability is seamless and easy.  As the organization needs to add switches to scale for a growing number of server and storage appliances, they need only plug the switches into the Plexxi network with one or two cables.  Configuration is automated with zero-touch and the network is self-discovering and self-documenting.

Through Plexxi Control, both the physical and logical network topology is fully documented and understood.  Based on the needs of individual data and application workloads, the Dynamic Fitting Engine can, in real time, change the network topology to deliver the resources and capacity required.  As an example, when a vMotion (movement of a workload from one location to another) is about to occur, Plexxi Connect can dynamically render a new network topology with a dedicated path providing the required bandwidth to ensure the vMotion happens quickly and successfully while ensuring it is not disruptive to the other critical application workloads in the data center.  A second, and very important example, is the “multi-personality” capability of the Plexxi network.  During the day, the topology is optimized to support the virtual applications and VDI environment, yet at night, when the VDI environment is dormant, the network topology dynamically changes to optimize storage and data analytics traffic.  This allows them to elastically and fully utilizes the total available pool of network capacity, without having the re-cable their physical network.  In addition, Plexxi Control delivers centralized management, control and visualization for their agile data center environment.

Plexxi Connect enables workflow integration with VMware (vSphere) and OpenStack, so the process of configuring and creating new VMs dynamically configures the network at the same time!  This integration delivers unprecedented operational efficiency and increased agility.  The need for inter-department trouble tickets to accomplish virtualized workload deployments and network provisioning is eliminated.

This enterprise set out to find a solution that was easier to deploy and manage, took less rack space, used less power and required less cooling.  They wanted a solution with the performance and scalability to meet their current and future virtualized application and storage needs.  They wanted to increase agility, improve operational efficiency, fully utilize an elastic pool of network resources and deliver workflow integration between their physical and virtual environments.  In Plexxi, they found a solution that accommodated all their needs of today and those of the future.

In my next blog, I’ll share the details of another customer use case from my travels. I welcome your comments and thoughts.

The post The Next Generation Agile Data Center and the Birth of A Dynamic Network appeared first on Plexxi.

Read the original blog entry...

More Stories By Richard Napolitano

Successfully leading a technology company requires a strong sense of what the market needs and an uncanny ability to build it. Plexxi CEO Rich Napolitano combines the vision required to disrupt with the experience necessary to succeed. He has spent more than 30 years in the IT industry, most recently as the head of EMC’s Unified Storage Division, where he was responsible for more than $4 billion in annual revenue and 2,000 employees.

Prior to EMC, Rich has led both sales and engineering, and acted as both founder and venture capitalist across companies like Sun Microsystems, Pirus Networks, and Alchemy Angels. Rich holds a BS degree in Computer Science from the State University, Stony Brook, NY, and owns two U.S. patents.