Now I’m a cynic by nature but there seems to be a fundamental transformation in progress here. Why? Because legacy data center networking equipment and operational processes are a mismatch for massive data center scale and dynamic cloud computing applications. ESG calls this imbalance data center networking discontinuity. The solution to this problem is fairly logical: Cloud platforms and server virtualization use software to turn hardware into a virtual platform. SDN, OpenFlow, DVNI, and even VMware VXLAN take the same approach.

To paraphrase the Monkees, I’m an SDN believer. Networks have to become virtual platforms that gracefully interoperate with cloud platforms like OpenStack. Provisioning new devices needs to be based upon a standard policy-based publish-and-subscribe model. Traffic engineering and security rules need drag-and-drop simplicity.

Pressing requirements, wide SDN adoption across the industry, and a new wave of engineering innovation will lead to an accelerated technology refresh cycle over the next few years. Yes, this has the potential to impact the networking status quo but I am reluctant to include hyperbolic terms like “game-changer” and “revolutionary” into my analysis for several reasons:

1. Network service providers and cloud computing service vendors are feeling the effect of data center networking discontinuity most acutely so it is logical for them to experiment with cutting edge technologies. Juniper proved in the 1990s that service providers are happy to take an alternative path if they see the potential to lower operating costs or increase ARPU so Nicira’s business model is spot on. By moving networking functionality to a virtual layer, Nicira gave itself a go-to-market advantage as it does not advocate a “rip-and-replace” hardware strategy (although it does turn feature-rich switches into an expensive transport layer). Still service providers feel the pain and have the motivation to move quickly.
2. Enterprises are not service providers and typically don’t have a team of network engineers to throw at a brand new technology. As a proof point, enterprise networking staff is finally comfortable with server virtualization although virtual switches retain the role of basic provisioning and access. Enterprises want a smooth migration path that addresses requirements and adds benefits over time without disrupting business operations.
3. What about software? This market will go a few ways: A new vendor could become the VMware of virtual networking by gaining a rapidly-growing market presence. With this position, it could become the networking software nexus with published APIs, SDKs, development partners etc. Alternatively, virtual network platforms could be based upon open standards or even open source a la Linux. In lieu of either of these models however, virtual network software becomes a proprietary game. Given Cisco’s market share, Cisco could turn Nexus into Windows and networking feature/functionality into Office. Regardless of the model, SDN will grow as a function of the software development community supporting it and that hasn’t happened yet.

The SDN market is very exciting but remains immature. I have no doubt that enterprise data center networks circa 2017 will look very different from the device-centric, manual process-driven model of today. How we get from here to there is a bit more difficult to forecast.

You can read Jon’s other blog entries at Insecure About Security.

via Riverbed evolves its Steelhead product family to better address performance, consolidation needs of global organisations | ITWeb.

Granted, information about the company has been around for a while, including an article in the New York Times last fall describing how NTT, in the wake of the Tsunami disaster, has been able to effectively leverage Nicira to maintain availability in the midst of rolling data center blackouts.  According to Nicira, other large service providers are also enamored with this technology, citing AT&T, eBay and Rackspace as customers as well. Nicira also cites a large enterprise customer, Fidelity, among those that have adopted its technology.

Why have these very large and well known service providers selected to work with Nicira? Simply put, because Nicira is helping them solve a very large, and fairly well known problem – network complexity in rapidly scaling, highly dynamic virtual environments. My colleague Jon Oltsik  has defined this as Data Center Network Discontinuity. As organizations continue to consolidate data centers, drive increased use of server virtualization technology, and increase their ability to react more quickly to business needs, legacy network solutions will reach a breaking point. Something needs to change in order to drive new and sustainable growth.  Organizations will need to create a virtual network platform to fully abstract the physical layer from the logical one, instead of an even bigger maze of individual boxes.

OpenFlow and Software Defined Networking hold a lot of promise in this regard. Many companies, including HP and IBM/NEC, are leveraging this technology and have recently made some big announcements, while other companies like Arista, BigSwitch, and Juniper have been pursuing their own implementations. Nicira hopes its model will take network virtualization to a whole new level. By placing the intelligence in the virtual switch and leveraging an external controller, Nicira hopes to virtualize (and commoditize) the underlying physical network.

By taking this approach, Nicira proves that it has no hardware agenda and therefore organizations do not have to rip and replace any equipment. It can be deployed in conjunction with existing technology from any vendor. As the customer list suggests, large service providers with cloud environments will receive the most benefit, but those enterprises that have committed to and deployed their own private clouds could take advantage of this technology as well, like Fidelity has done. As more enterprises consolidate data centers and mature their server virtualization environments, I would expect to see more enterprise logos pop up on its website, if the technology works as advertised with its initial customers.

You can read Bob’s other blog entries at Data Center Continuum.

To assess these issues, ESG asked 280 networking professionals to respond to questions in areas including:

• Data center strategy
  • How many data centers do large organizations have today?
  • Are large organizations consolidating data centers?  If so, how many data centers do they hope to eliminate?  Are they consolidating discrete business unit data centers into multi-tenant mega-centers?
  • Are they extending applications or IT operations across multiple data centers?  If so, which specific applications and IT operations activities are being extended?
  • Is server virtualization a substantial component of their data center strategy?  How will the use of server virtualization change in the future?
• Data center networking environment
  • What types of data center networking architectures are in place today?
  • How are networks segmented?  Why is some segmentation done at Layer 2 and some at Layer 3?
  • What is the standard server-to-network interface today?  How will this change in the future?
  • What is the standard data center network cabling infrastructure today?  How will this change in the future?
  • How is networking equipment provisioned and managed?
  • What is the role of virtual switches today?  How will this change?
  • What is the impact of web applications on data center networks?
• Data center networking operations challenges
  • What types of network operations problems do large organizations face today?
  • What are the biggest network operations challenges?
  • Which network operations activities are most costly and time consuming?
• Data center networking adaptation
  • How is the networking organization changing to adjust to data center scaling and operational requirements?
  • Are large organizations adopting new data center networking technologies?  If so, which ones?
  • Are enterprises implementing converged data center networks (i.e. common network for storage and data traffic)?  If so, why and how?
  • What changes are taking place in order to support more server-to-server web application traffic within the data center?
• Data center networking strategy
  • Which data center networking technology innovations are users aware of?
  • Are enterprises interested in data center network fabric architecture?  Will they implement data center network fabrics over time?

For more information on the contents and findings of this report, please download the executive summary below.

ESG Research Report Data Center Networking Trends Executive Summary

Click the picture for a PDF of the Data Center Networking Infographic.

Although most overall IT budgets are edging up slightly, network budget growth is much more robust. Almost 60% of respondents (58%) report that they will be increasing their network budgets in 2012. The larger the organization(enterprises with 1,000 or more employees), the greater the increase. In fact, 21% of enterprises expect to increase spend by 8% or more.

Where will this money be spent? According to our research it will be spent on:

• Network security. As the network environments scale, organizations will require their security services to scale with them and be more tightly integrated with the network. For more on security spending see Jon Oltsik’s security blog.
• Network management. The key to managing more with less will be better network management tools. However, with only so many span ports to connect to, organizations will need to find solutions to help them scale their network management/monitoring coverage for the whole environment. This could bode well for companies like Anue, Gigamon, Netoptics, and VSS that will be instrumental in providing visibility across massively scalable networks. Especially for those 10 GbE networks. This management at scale may also open the door for SDN/OpenFlow vendors to highlight the virtues of their solutions.
• WAN optimization. It’s not just for the troublesome remote office anymore. As organizations continue to consolidate data centers, connect to cloud and SaaS providers, and try to deliver a solid user experience for remote workers, we expect more organizations will be taking an enterprise wide approach to optimizing their connections. Especially as latency sensitive applications like video, desktop virtualization and VoIP (see next bullet) continue to proliferate in the data center. Established WAN optimization vendors like Blue Coat, Cisco, Citrix, Riverbed, SilverPeak, and others have already started to adapt their solutions to meet these emerging needs.
• VoIP. One third of the respondents indicated that they would make investments in VoIP solutions. With a rapidly changing and increasingly remote workforce, VoIP solutions provide greater flexibility options. As networks continue to converge, we expect this will also include video and even desktop virtualization as well. So organizations would do well to look for both solutions.
• Headcount. ESG found that 36% of organizations plan to hire additional network staff – fortunately, most believe there are plenty to be found. The only area that showed a higher increase in adding more people was security.

You can read Bob’s other blog entries at Data Center Continuum.

To me, this is bigger than just a press release and some joint marketing programs. Here’s why:

1. IBM and NEC are moving OpenFlow beyond academic labs and cloud computing theory, taking their joint solution to enterprise data centers. Yes, enterprises need to be educated on SDN and its benefits, but the use case for OpenFlow is certainly there since legacy networks can’t keep up with growing data scale or virtual server mobility.
2. While the headline may be OpenFlow, it’s really all about software. Mainframes became virtual computing platforms in the 1970s and Intel servers did the same with server virtualization technology from Citrix, Microsoft, Red Hat, and VMware. The next step is cloud computing which is intended to virtualize the whole IT infrastructure enchilada but static proprietary networks just don’t play well in this arena.
3. You have to give NEC credit for recognizing the software-centric opportunity around OpenFlow and bringing a quality controller to market. NEC could become the standard glue of a heterogeneous OpenFlow network over time.
4. When HP purchased 3Com, a lot of people had IBM reacting with an acquisition of Brocade or Juniper. With SDN/OpenFlow, IBM can create a data center fabric out of access switches. Between OpenFlow and existing partnerships, I can’t see IBM making a big networking acquisition anytime soon.
5. For those of us who’ve been around the industry for a while, it is certainly ironic to see IBM taking a leadership position in networking. I know I’m showing my age, but it doesn’t seem like that long ago that IBM was pushing Token Ring and SNA.
6. Personally, I don’t see SDN and OpenFlow as a threat to Cisco. In fact, Cisco could build OpenFlow software with IOS/Nexus intelligence and integration as sort of a dual path strategy. If I’ve learned anything about the network industry it is this: Never (and I mean never) count Cisco out when it comes to networking.

You can read Jon’s other blog entries at Insecure About Security.

A similar binary situation is developing with data center networks. On the one hand, the environment is going through some radical changes. According to ESG Research:

• 63% of enterprise organizations are consolidating or have already consolidated data centers. On average, these firms will eliminate 25% to 50% of their data centers. Nearly half of organizations will consolidate data centers belonging to independent business units into multi-tenant facilities.
• About 40% of organizations are extending applications or moving virtual servers across geographically dispersed data centers. In other words, web applications and server virtualization are creating a virtual data center platform across multiple physical facilities.
• Almost all enterprises are using server virtualization today with aggressive future plans for more VM workloads, more VMs per physical server, etc.
• Large organizations anticipate steady growth in the number of physical/virtual devices per data center, the number of VLANs per data center, and the number of IP subnets per data center.

Clearly the data center environment is changing, which begs the question: Are data center networks adapting? Yes, but the ESG data indicates that the slow pace of change is causing some major problems. When asked to identify network operations challenges, large organizations pointed to things like too many manual processes, time consuming network provisioning and configuration, and organizational problems between the networking team and other functional IT groups. Network security is a mess and network performance is a constantly moving target. Yikes!

The data center networking dinosaur is adapting but we are rapidly approaching a breaking point. Yes, the networking industry is working diligently to bridge this gap, the ESG data points to an inevitable inflection point coming sooner than you think.

I will continue to blog about the ESG data center networking research as the report will be published quite soon. A few parting thoughts here:

1. Expect changes in every aspect of data center network: NIC cards, cabling, virtual switches, data center fabrics, etc. Networking vendors must realize that these are a lot of simultaneous change that even the most sophisticated IT shops will have difficulty understanding.
2. If anyone was still unclear about why Cisco went into the server business, my first point should solve this riddle. The ESG data also indicates that Cisco is succeeding but more on this later.
3. While the jury is still out on OpenFlow, there is no question that the future of provisioning, management, and control planes will be software-based. The network must become a virtual platform a la VMware, Xen, KVM, etc. It is likely that SDN, OpenFlow, and vendor support from companies like BigSwitch, Brocade, HP, IBM, Juniper, and NEC will get a lot of attention in 2012.
4. If you believe #3, Arista’s strategy looks increasingly intelligent.
5. While pure-play cloud computing initiatives are still rare, many enterprise organizations are extending applications, moving workloads, and using global load balancing across multiple data centers. Seems like the on-ramp to cloud computing to me. Once again, the network must adapt to location-independence — or die.

More soon.

You can read Jon’s other blog entries at Insecure About Security.

via Energy savings in the cloud – The Times of India.

That’s where the ESG Lab validation of NEC’s ProgrammableFlow comes in. NEC has taken the OpenFlow standard and created a powerful software driven networking solution that allows IT professionals true dynamic provisioning of network resources to support application requirements. In today’s virtual server environments this is an exciting proposition as now your network can be as nimble as your virtual deployment.

I won’t go into the details of the full report here, but I’m looking forward to seeing more of these solutions developed for the commercial market. It’s easy to see how this new architecture can reap huge benefits for both public and private clouds.

Click here to read the full lab report and see how ProgrammableFlow can be a viable solution for today’s enterprise network.