Network virtualization is an emerging technology that enables datacenter managers to ease VM (virtual machine) provisioning challenges, improve automation, and support the scalability demands of large datacenters.
Network virtualization (NV) is in its very early stages of adoption and encompasses several competing standards, a large number of technology suppliers, and multiple open source options.
Key questions about the NV market include:
How rapidly will service providers and large enterprises adopt NV technology in their datacenters?
Does the NV develop into a distinct market (VMWare's view) or become a feature of broader software-defined networking (SDN) deployments (Cisco's view)?
Who will be the leading buyers of NV technology? Will sales be driven by network administrators, server administrators, or datacenter managers?
Definition of network virtualization
Network virtualization uses tunnels to create virtual networks on top of traditional (Ethernet) physical networks. Leveraging protocols such as VXLAN or NVGRE, NV provides Layer 3 tunneling which separates the virtual network from the underlying physical network hardware, thus enabling provisioning of virtual networks and accompanying services between servers in the datacenter.
Benefits of network virtualization
Implementation of network virtualization provides a number of key benefits for datacenter managers, including:
Scalability -- the ability to scale beyond vLAN limits to support hyper-scale datacenter network requirements.
Multi-tenancy -- each application or tenancy can have their own network and security policy via NV traffic isolation.
Ease of implementation -- no need to change the underlying physical network.
Improved performance of VM-to-VM traffic within the same server or rack.
Network virtualization is offered by a significant number of technology suppliers. Each supplier has as different take on network virtualization with accompanying strengths, weaknesses, protocols, and pricing options.
VMWare offers its NSX product (via its Nicira acquisition) that leverages the VXLAN protocol to provide network virtualization.
Microsoft bundles network virtualization functionality (using the NVGRE protocol) in its Hyper-V product.
Nuage, a division of Alcatel-Lucent, offers the Virtualized Services Platform (VSP) network virtualization technology. Nuage offers a number of pricing options: per server or per VM, subscription or license.
PlumGrid offers its IO Visor technology, which enables the creation of virtual domains that replicate the physical network.
Juniper offers network virtualization via its Contrail by license or subscription. Juniper also offers Contrail via free open source software.
Midokura offers MidoNet, which virtualizes the network to let OpenStack and CloudStack create and manage virtual networks.
There are a number of open source options for network virtualization, including OpenDayLight and Open vSwitch. Other NV technology suppliers include: ADARA, Big Switch, IBM, NEC, Cisco, Redhat, HP, and Arista.
NV market variables
Forecasting the market for network virtualization market involves a number of key variables, including:
Pricing models: What are the NV pricing models (by VM, server, or socket) and by license (one-time fee) or monthly subscription? A key variable here is the potential for NV functionality to be built into either the hypervisor (e.g. Microsoft) or included in the network infrastructure costs (e.g. Cisco).
Growth of cloud service provider infrastructure: The growth of new hyper-scale datacenter designs will fuel the requirements for NV functionality. Key metrics include the growth of OpenStack and Open Compute deployments.
Impact of open source: The SDN community is creating a lot of free, open source software, including OpenDayLight and Open vSwitch, which has the potential to offer NV functionality. Faster adoption of SDN open source would push NV spending from software towards support and professional services.
Current and future market adoption
Current adoption of NV technology is in its very early phases with a few leading-edge organizations in production use and more communications service providers and enterprises in the pilot phases. Running NV software as an overlay to existing network infrastructure provides a relatively easy way to alleviate VM networking challenges. As a result, I expect NV adoption to increase strongly during the 2015-2016 timeframe. The future adoption of NV technology will be dependent on a number of key variables, including pricing, standards, and the impact of open source. I forecast that spending on the SDN software market (of which NV is a subset) will reach nearly $1.2 billion worldwide by 2018.
mhhf1ve, User Rank: Light Sabre 2/27/2014 | 5:20:13 PM
Large, conservative organizations... will adopt NV when it's mature As you noted, NV is in its early stages. So large companies will likely sit back a bit to see how standards pan out and make sure the risk isn't too high to deploy. I'd guess that part of the equation will be assessing the maturity of the open source projects -- because as soon as there is a quorum of developers working on stable open source projects, then companies will start to feel comfortable relying on NV in their datacenters.
sam masud, User Rank: Light Sabre 2/21/2014 | 11:15:31 AM
Re: Timing It's interesting that most of the focus seems to be on NV based on encapsulation/tunneling as opposed to implementing NV via SDN flow tables--interesting because the network overlay approach doesn't (from what I understand) allow bandwidth reservation by network overlays.
Timing oIt sunds like there is not much more time for NV to develop as a distinct market. With SDN on the horizon, won't a lot of datacenter operators just tackle all of the above at the same time as part of upgrading to a next-generation datacenter architecture?
Light Reading founder Steve Saunders recently visited the University of North Carolina Charlotte (UNCC) where Cisco's Tetration application is providing data center analytics, simplifying SDN, helping with cloud migration and overseeing white-list security policy.
A CSP's digital transformation involves so much more than technology. Crucial – and often most challenging – is the cultural transformation that goes along with it. As Sigma's Chief Technology Officer, Catherine Michel has extensive experience with technology as she leads the company's entire product portfolio and strategy. But she's also no stranger to merging technology and culture, having taken a company — Tribold — from inception to acquisition (by Sigma in 2013), and she continues to advise service providers on how to drive their own transformations. This impressive female leader and vocal advocate for other women in the industry will join Women in Comms for a live radio show to discuss all things digital transformation, including the cultural transformation that goes along with it.