Virtualization and Cloud executives share their predictions for 2016. Read them in this 8th Annual VMblog.com series exclusive.
Contributed by Bruce Davie, CTO for Networking at VMware
Network Virtualization: Mainstream in 2015, Expanding Reach in 2016
Once again it's time to look back on the last year, see what happened
in our industry, and try to make some predictions for the coming year.
Let's start by seeing how last year's predictions fared.
One of my main points last year was that 2015 would be a year of
operationalization and production deployment of network virtualization.
Indeed, 2015 saw hundreds of production deployments of network
virtualization. As I write, over 200 customers are running VMware NSX in
production data centers, with more being added every week.
With the rapid uptake of network virtualization in production
settings, our team made a big push on operational aspects of NSX. On the
product side, NSX 6.2 shipped in August with a slew of new features
focused on operations, such as traceflow
and a centralized CLI. We also invested heavily in documenting the best
practices for operationalizing NSX in production. By surveying some of
our successful production customers, we developed a set of guidelines
and best practices for network virtualization deployments, which we
documented and presented at VMworld.
Two observations stand out as we look at customer adoption of NSX in
2015. The first is the diversity of customers, in terms of both size and
industry. This is an indication of the maturing of the market for this
technology. In contrast to the early days, customers don't need to be
especially large or sophisticated to see the benefits of network
virtualization, or to put the technology into production. While we still
see lots of interest from our traditional large, technology-focused
customers, adoption is clearly spreading across industry segments such
as healthcare, retail, the public sector, and many others. Additionally,
there are plenty of smaller customers in the mix with large enterprises
and service providers.
The second observation is that no single use case is dominating
adoption of network virtualization - in fact, the breadth of use cases
continues to increase. While the emergence of microsegmentation as a use
case definitely increased customer interest in NSX and continues to be
important, customer deployments are spread across many different use
cases such as agility and automation, service insertion, and
multi-data-center applications such as disaster recovery. Increasingly
we are seeing customers tackling multiple use cases in a single
deployment. Overall this breadth of usage points to the general-purpose
nature of network virtualization technology.
Now, what about 2016? Aside from confidently predicting more
customers, deployments, and use cases, there is actually one other
notable trend that should emerge in 2016. That is an increase in the
range of endpoints that can be managed by NSX (a phenomenon that our
marketing team is calling "NSX Everywhere"). This is a natural extension
to the development of NSX over the past three years. From the early
days we've had support for workloads running on a range of hypervisors
(ESX, KVM, Xen), and we've been able to extend virtual networks to physical workloads as well. By integrating NSX with AirWatch,
we've been able to extend the security capabilities of
microsegmentation to applications running on mobile devices. We've
started to use microsegmentation to improve the security of virtual
desktops.
In 2016 we'll see this ability to provide networking and security
services to a range of endpoints move to another level. The set of
endpoints that NSX can manage will extend to containers and public cloud
workloads. We'll also see NSX extending out to branch offices as
Software Defined WAN (SD-WAN) solutions take root.
We actually demonstrated
the extension of networking and security controls to containers and
public cloud workloads at VMworld 2015. We'll be working to productize
these demos, and it's becoming clear that these extensions to the reach
of NSX will be important in 2016. A common theme among customers I've
talked to is that they like the security and visibility features that
NSX provides, and would love to have those same features available for
public cloud workloads. Recognizing that for at least some of their end
users, public clouds will be the chosen venue for a workload to run, IT
managers want to have a consistent view of networking and security
policy. They also want to maintain that consistency even if a workload
at some point moves from one public cloud to another, or moves back to
on-premise deployment. Meeting this requirement will be the objective as
we expand NSX into public cloud environments. A similar desire for
consistency in networking and security policies will drive the
extensions to support containers as first-class endpoints for NSX.
Finally, it’s clear that SDN is spreading out of the data center and
into the WAN. This is not the time to rehash arguments of what is and is
not SDN, but let’s just say that logically centralized control of a
distributed data plane, which has worked so well in enabling network
virtualization to succeed, is now also being successfully applied to
wide-area networking problems. There is a natural opportunity here for
NSX to extend its reach out to branch offices. While the details of
various SD-WAN implementations vary, all of them provide some mechanism
for automatically building overlay tunnels among remote sites and data
centers. These tunnels can readily be connected into a virtual network
managed by NSX, at which point NSX can start to provide networking and
security services to branch endpoints as well. Microsegmentation can
extend out to the branch, separating the traffic of retail customers
from that of internal IT operations, for example. We are already heading
down this path with some of our customers, and we expect to see a lot
more of this usage of network virtualization in 2016.
So, it continues to be an exciting time for network virtualization.
Adoption of the technology will increase, and we’ll see still more
breadth of customer types and use cases. Perhaps most exciting is that
we’re moving well beyond our traditional “sweet spot” of delivering
networking services to on-premise virtualized workloads, as we expand
the reach of NSX to everything from handsets to public clouds to the
software-defined WAN.
##
About the Author
Bruce Davie is CTO for Networking at VMware, and a Principal Engineer in
the Networking and Security BU. He joined VMware as part of the Nicira
acquisition, and focuses on network virtualization. He has over 25 years
of networking industry experience, and was a Cisco Fellow prior to
joining Nicira. At Cisco, he worked closely with leading service
providers to enhance the capabilities of their networks. He led the team
that developed multi-protocol label switching (MPLS) and contributed to
the standards on IP quality of service. He has written over a dozen
Internet RFCs and several networking textbooks. Bruce received his Ph.
D. in computer science from the University of Edinburgh in 1988 and is
an ACM Fellow.
