Philip Sellers, who writes the Tech Talk blog, recently wrote an in-depth review of the NextIO vNET Maestro, our new top-of-rack IO virtualization solution.  Philip is a systems administrator for America’s largest telephone cooperative, Horry Telephone Cooperative, in Conway, SC.  His primary focus is on Microsoft operating systems and infrastructure, VMware virtualization, and blade servers for the co-op.

From his post:

Advantages:  The advantages that I see with the technology are very similar to the benefits I initially show with HP Virtual Connect, although a little different since it is with more industry standard hardware.  Some of the advantages include:

• Reduced need of ports for Fiber Channel and Network, which is really useful when few applications can consume the capacity of an 8Gb fiber channel port or a 10Gb Ethernet port.   As with virtualization in its first phase which sought to increase utilization of processor and memory in hosts, this technology can be used to increase utilization of under-utilized connectivity.
• Less costly upgrades to the newest technology allowed by adding a single PCI card of a specific technology and then sharing the new features to the backend rack mount systems.
• Server to server traffic not interacting with systems outside of the vNet’s domain does not need to leave the vNet device and can travel between systems at PCIe native speeds.
• Very intuitive and easy to learn interface to provision interfaces to backend nodes.
• The special PCI riser cards are pass-through modules with no firmware or intelligence to patch and maintain.
• Interconnect cards are industry standard PCI cards – nothing special or proprietary.
• PCI traffic passes through the vNet unit even without the management module in service, meaning that a firmware upgrade or other outage in the module would not cause downtime.

You can check out the entire post here for more detail:

http://tech.philipsellers.com/2011/11/30/nextio-releases-the-vnet-io-maestro-a-san-and-network-convergence-device/

Thanks for the review Philip!

I/O Virtualization provides agility to solve tough virtualization projects

In 2011, IT departments have been asked to continually do more with less:  deploy on-demand clouds and enterprise wide VDI, virtualize mission critical applications, upgrade to 10GbE, and implement ever more complex tiered storage solutions.  At the same time, IT budgets have remained flat or even decreased.  Architects are realizing that current hardware designs are inefficient at deploying I/O resources to keep up with the agility and reliability required for virtual desktop and cloud deployments.  Current designs are both expensive (have you priced a 10GbE uplift lately?!?) and inflexible.  Much like server virtualization 8 to 10 years ago, I/O virtualization is a disruptive technology that improves reliability, consolidation, and flexibility, while ultimately lowering IT costs.

What do I expect with regards to I/O virtualization in 2012?  Read on:

I/O virtualization provides the key to cloud agility

I think we can agree that for many enterprises the easy stuff has been virtualized.  Now admins are tackling much harder problems:

• Virtualizing database servers and mission critical and high performance apps
• VDI, which has been evaluated to death over the last 3 to 4 years, must now be deployed in production. End users are sensitive to the performance of their desktop apps and IT staff will hear about any deficiencies.
• Internal customers are no longer satisfied with the “one size fits all” VM templates deployed in the cloud – don’t box me in!

To keep up, IT staff will have no choice but to invest heavily in their current architectures or look to new technologies such as I/O virtualization.

Much like server architecture before virtualization, currently I/O resources are inefficiently deployed, inflexible, and expensive, generally with a lot of excess capacity.  I/O virtualization is the separation of I/O (Ethernet, Fiber Channel, FCoE, etc.) from the compute resources (CPU and Memory) so that the I/O resources can be consolidated, managed, and deployed most effectively.

I/O virtualization provides both cost savings and flexibility, with the agility to match I/O resources to workload requirements with the same ease as provisioning servers.  The hardware limitation of the host is removed so that staff can easily scale I/O up or down according to the VM workload (as well as bill customers for the I/O resources consumed).  In addition, GPUs and SSDs can be added for specific high demand workloads.

Eliminate vendor lock-in and reduce hardware costs

Proprietary architectures such as blade chassis’ or interconnect fabric can be very effective for certain virtualization projects, but that effectiveness comes at a price.  Many enterprises will be looking for ways to match the flexibility of their hardware architecture to the flexibility of their cloud services.  This includes technologies that scale up easily and cheaply, accommodate a variety of vendors, and do not force them to make long term bets on the viability of certain architectures to meet their processing needs 4 or 5 years out.

While third party cloud providers are stepping in to provide scalable resources on an as-needed basis, I/O virtualization provides much needed flexibility and cost savings for dedicated environments.  I/O virtualization is based on industry standards such as PCIe so that it can accommodate compute nodes from a variety of vendors – pick your server according to availability, cost, and performance, not on a blade chassis purchased years ago.  IT staff can easily deploy new technologies to take advantage of cost savings or performance improvements.  I/O virtualization provides a cost effective way to deploy 10GbE or Fiber Channel in the datacenter.

In 2012, IOV will help enterprises tackle some of their more difficult virtualization projects.