Welcome!

Eclipse Authors: RealWire News Distribution, Michael Bushong, Matthew Lobas, Mark R. Hinkle, Shelly Palmer

Blog Feed Post

Plexxi Pulse – Vote for OpenStack

The OpenStack Summit session voting is officially open, and Plexxi has two sessions in the running. Check out sessions from Nils Swart, The Future of OpenStack Networking, and Derick Winkworth, Group Policies for Neutron and evolving the abstraction model to merge with OpenDaylight, and get your votes in. In our video of the week, Dan Backman explains how the Plexxi Pod Switch Interconnect has increased the size of our product portfolio and looks at the differences between the switching platforms. Check out our video of the week and a few of my reads in the Plexxi Pulse – enjoy!

Eric Krapf, contributor to No Jitter, discusses how the communications industry is embracing SDN more and more as evident by Microsoft and HP’s use of a Lync API that can connect the communications servers with the controllers in an SDN architecture. This article has a great discussion, even though it’s not surprising that SDN is relevant in communications. If you think of communications as just another application of the network, then the idea that SDN will enable app-network exchanges is a natural extension of the technology. The issue is that people don’t frequently think of SDN as enabling app-network collaboration. It has gotten a fairly narrow definition around controllers and OpenFlow, which misses the point of abstractions and workload delegation. This article provides a very practical example of what can be done and highlights how SDN doesn’t need another 3 years to make an impact.

Jude Chao, editor at Enterprise Networking Planet, says more mergers and acquisitions will occur in the networking industry in 2014 according to a report released by PricewaterhouseCoopers (PwC) this week. Jude says SDN is a major factor driving this industry shift. I agree with the underlying premise – that there will be consolidation. A small number of networking hardware and software players will get acquired and several startups will fail to pick up traction and lose out on subsequent rounds of funding. I also believe a few smaller vendors will break out as independent players. Interestingly, the action will be in the periphery as well. Analytics, monitoring tools, data collection, DevOps, and even point solutions for niche applications will do well, and a round of change for the VARs will occur. Deploying SDN will mean a shift in business models, and not all VARs will make the leap. Much of this industry shift is not necessarily tied to commodity hardware, because the underlying being hardware cheap is not that big of a driver for industry consolidation. Differentiation already exists in the software today, even if pricing favors hardware.

The Register’s Jack Clark discusses the “seismic shifts” in networking and focuses on a quote from AT&T’s head of technology and network operations, John Donovan, in an interview with the Wall Street Journal, saying that the telco’s Supplier Domain Program 2.0 saves money. The networking industry has had good margins for a long time. This isn’t because of the hardware being intelligent or not. It’s because there hasn’t been much competition. The thousands of features that get deployed mean that the number of functionally equivalent devices for a particular spot in the network is small. With little competition, pricing stays high. Things like SDN are important for two reasons: first, they reset the architecture to some extent, which reduces the power of all those legacy features, and second, it helps automate workflow. The first point increases competition and drives prices down. The second addresses the bigger cost issue: managing all the devices. Ultimately, Cisco will drop their prices as competition heats up. They will be a player in the future of AT&T, just maybe not to the same extent they are today. But the real battle is going to be over long-term OpEx reduction. Merely making a cheaper switch doesn’t address that. If AT&T just wanted the same network they have today at lower prices, they would put pricing pressure on Cisco. This is about something much larger.

Blogger Ethan Banks contributed an article in Network Computing about how Ethernet switches and the purchasing process have changed in the last few years. He says buyers today “must learn a variety of technical nuances that set switches apart from one another, match those capabilities to their organization’s needs, and then move ahead to a purchase.” After reading this post, I wonder what the role of off-box capabilities will be in Ethernet switch selection in the future. SDN is about workflow automation. People interested in that will also key in on things like orchestration and DevOps. It could be that on-box support for what ultimately ends up being off-box functionality will matter more. I only mention this because I suspect that people will need to broaden their selection criteria beyond the box to include things like Puppet or Chef integration or even OpenDaylight support. It will take a confusing process and make it potentially even tougher in the short term. Those customers with a more solid grasp of current and future strategy will be in a better position to make these types of decisions.

ReadWrite contributor Jonathan Crane explains how the IT department will make important strides toward driving innovation and growth in 2014. Jonathan analyzes a recent Gartner that predicted numerous developments that will greatly impact the IT function across mobile device management, hybrid cloud integration and SDN. One of the things that becomes necessary in an infrastructure environment that is orchestrate as a whole and in support of the applications is the expression of application requirements in application terms. Basically, to operationalize things, someone has to be able to capture what is important across the infrastructure. This cannot be specified in networking language or compute language. It has to be expressed relative to the application. The various infrastructure systems then need to translate the requirements into underlying behavior. I mention this because someone has to own the application abstraction. That would seem to fit with your definition of the OM. And then the OM would translate (or facilitate the translation of) the application requirements into underlying configuration primitives. This obviously has to be done through data models and APIs; a manual translation would leave us where we are today. The question people need to be asking then is: who is defining the abstractions? And what tools do I need to use them? This is where open source projects like OpenStack and OpenDaylight come into play. Anyone who is in an OM role (or wants to be) needs to be looking at these projects very closely to understand how to intersect their IT operations with the availability of management frameworks and controller architectures.

Mitch Wagner at Light Reading says according to a Forrester analyst, Cisco customers can stop purchasing the company’s switches and Cisco will still prosper. There will always be people who predict the demise of the incumbent. That might be hyperbole, but there will certainly be headwinds. I don’t think Cisco is incapable of executing against an SDN strategy. They have proven they can develop products, and when in doubt, they have mastered the strategic acquisition. SDN, however, is a new architecture. The new architecture reduces the need for the tomes of legacy features that have made it exceedingly difficult to get off the Cisco drug. With a new architecture, you get a more level playing field with lower barriers to entry. It’s the increased competition that will whittle away share. Will it be 20 or 30 points? Probably not, but you could see a significant share movement over the next 3-5 years.

Tom Hollingsworth, the Networking Nerd, says SDN vendors are creating an event horizon, which is a boundary beyond which events no longer affect observers or the point of no return for things falling into a black hole. If SDN enables bidirectional communication between the apps and the network, it stands to reason that you would begin to architect each of them differently. Obviously you must start with making it possible; no one will change anything if there is no support for it, but you create applications that take advantage of network information. Imagine massive data replication jobs. If they are not time critical, you could schedule them and create pipes across the network. You could serve content from caches that were less congested. You could do things like variable bit rate for mobile connections that are shifting from 3G to Edge and back to LTE on a train ride. Ultimately, I agree with the premise of this post. I don’t think the future is overlays that are completely agnostic to the underlying network. I think there will be a desire to pin the overlays to the physical infrastructure and allow for the dynamic optimization of the physical transport to suit whatever is happening on the overlay.

 

The post Plexxi Pulse – Vote for OpenStack appeared first on Plexxi.

Read the original blog entry...

More Stories By Michael Bushong

The best marketing efforts leverage deep technology understanding with a highly-approachable means of communicating. Plexxi's Vice President of Marketing Michael Bushong has acquired these skills having spent 12 years at Juniper Networks where he led product management, product strategy and product marketing organizations for Juniper's flagship operating system, Junos. Michael spent the last several years at Juniper leading their SDN efforts across both service provider and enterprise markets. Prior to Juniper, Michael spent time at database supplier Sybase, and ASIC design tool companies Synopsis and Magma Design Automation. Michael's undergraduate work at the University of California Berkeley in advanced fluid mechanics and heat transfer lend new meaning to the marketing phrase "This isn't rocket science."