Analyst Report: SDN, NFV, and open source: the operator’s view

Credit: Thinkstock

1 Summary

Software-defined networking (SDN) and network functions virtualization (NFV) represent two of the more dramatic oncoming technology shifts in networking. Both will significantly alter network designs, deployments, operations, and future networking and computing systems. They also will determine supplier and operator success (or failure) over the next five to 10 years.

As has always been the case with successful networking technologies, industry standards and open systems will play a strong role in the timely widespread adoption and ultimate success of both SDN and NFV solutions. Open source is poised to play an even more critical role in delivering on the promise of standardized and open networking.

This great promise and potential impact begs two questions. First, “Where are SDN and NFV today?” And second, “What influence will open systems and open source have on the future of SDN and NFV?”

To find answers to these questions, in December 2013 Gigaom Research ran an extensive survey of 600 operators (300 enterprises and 300 service providers) in North America. Based on findings from that survey, this research report provides key insights into the current activity and future direction of SDN and NFV advancements as well as the development and deployment of open systems and open source within SDN and NFV environments.

Key highlights

  • SDN and NFV deployment timelines are extremely aggressive. While these timelines will certainly move out as real-world pressures (financial, technical, and organizational) force delays, these results indicate the high level of hope for SDN and NFV solutions.
  • Security continues to be a major challenge in networking. While SDN stands to solve many problems, improvements to the security posture of the world’s networks remains a primary driver across all networking advancements, including SDN and NFV. After security, key drivers include improved network service levels and lower operating and capital costs.
  • While much is made of the potential for SDN and NFV to optimize network spending, utilization, and service levels, near-term improvements in network operations are the primary focus for operators looking to deploy SDN and NFV solutions and take advantage of open source within their SDN and NFV environments.
  • The application of SDN and NFV solutions is varied – and surprising – across the respondents. As expected, the data center is a primary initial target for SDN and NFV solutions. However, for enterprise respondents, the wide area network (WAN) takes precedence over the data center. And likely reflecting the pressure of the bring-your-own-device (BYOD) movement, enterprises are targeting the wireless local area network LAN (WLAN) before the campus LAN, branch WAN, or metropolitan area network (MAN). For the service provider, the data center is far and away the No. 1 initial target. Interestingly, however, the LAN and the wireless WAN (WWAN) are targets two and three, respectively, for the service provider – not the WAN or the MAN.
  • Roadblocks that would push out the aggressive SDN and NFV deployment timelines planned by operators include migration costs, clear and consistent capabilities, and unproven performance and reliability. Interestingly, interoperability is viewed as the least of the concerns with SDN and NFV, perhaps because operators look at the close ties between SDN, NFV, open systems, and open source as a true saving grace, relieving them of the burden of heightened systems integration, which is the problem-resolution work that comes with an undoubtedly more mixed-vendor SDN and NFV environment in the future.
  • As expected, demand for industry standards and open systems within SDN and NFV solutions is strong. Freedom of choice and potential cost savings are obvious gains. Surprisingly, operators also expect open systems to deliver strong functionality. They are not simply looking for least-common-denominator functionality. They are looking for open systems to provide basic interoperability and leading-edge capabilities in areas that have traditionally challenged proprietary single-vendor solutions (e.g., administration, analytics, and security). Operators absolutely prefer open systems. They also demand that open systems add value beyond interoperability.
  • Open systems and open source are tightly linked in operators’ minds, indicating that open source is the preferred delivery method for standardized open SDN and NFV solutions. However, operators strongly prefer that open source be delivered by commercial suppliers. Here the operator looks to gain the best of both worlds – the freedom and functionality of open source combined with the proven delivery and support practices associated with commercial solutions.
  • Enterprise and service provider operators are closely aligned across all major areas of consideration and concern. They also are closely aligned in terms of deployment timelines. Among the service providers, SDN and NFV are also closely aligned in terms of deployment timelines, moving almost in lockstep over the next two years. The one area where enterprises and service providers differ is in their initial target for SDN and NFV deployments.
  • The perceived challenges to applying open source to SDN and NFV environments are also important. Operators see security and reliability as the biggest impediments to using open source when deploying SDN and NFV solutions. Does perception match reality? Are we seeing outdated open source perceptions at work as we did in the early days of OpenStack, Linux, and other successful open source projects? Given that today’s open source is conceived and developed by teams of operators and suppliers and increasingly delivered via proven commercial integration, test, packaging, deployment, and support models, perhaps in this instance operators are more wrong than right. Obviously the onus is on the open source projects to alter this perception. Alteration leads to acceleration of open source acceptance within SDN and NFV environments.

2 The state of the network

If networks were maximally efficient and effective, operators would not be looking to make the dramatic changes that SDN and NFV bring. What is driving the need for change?

Key findings: expected and unexpected

“The more things change, the more they remain the same” certainly applies to networking. Technology advances rapidly, and yet similar challenges continue year in and year out.

The following survey responses serve to highlight this “changing but same” state of the network:

  • Security has long established itself as the chief concern in the networking age. While these survey results may have been influenced upwards by the untimely and highly publicized 2013 holiday security breaches, security continues to demand operator attention – and budget.
  • While always hard to admit for operators, wider and more widespread cracks in the network are having a negative impact on network service levels. Whether from bandwidth shortages, device slowdowns, management shortcomings, increasing user and information demands, or pressured budgets (or likely some combination of all of the above), networks are showing signs of stress across many fronts.
  • While possibly influenced by the strong presence of budget and business overlords in the respondent mix (e.g., executives and managers), cost pressures continue to mount for both equipment and operations. While opex outweighs capex 3-to-1 or 4-to-1 (even 5-to-1) in networking and computing budgets, spending on capital equipment often comes in big waves – and all too often for most budget managers. This is likely why the capex argument remains very visible in SDN/NFV discussions despite the fact that SDN and NFV stand to save more absolute money on the operations side and drive more revenue on the business side.
  • While much of the attention on SDN and NFV focuses on programmability and provisioning, the figure below indicates that these two areas are of lesser concern with respect to today’s networks. Doing better at “operating now” seems to be far more the focus for operators than “optimizing new.” However, compared to past survey results, new technology adoption, service rollouts, and user/resource deployments are rising in importance. This is good news. Deploying new systems, services, sites, and users – more readily, accurately, and simply – will go a long way toward improving the areas of greatest concern (e.g., security, service levels, and capital and operational savings).

The big challenges: some old, some new

Security is well out in front as the primary concern of operators; 67 percent of respondents rank security vulnerabilities as a primary concern. All other concerns but one are cited by between 50 percent and 60 percent of respondents. A 7-percent point lead for security over the second biggest concern likely translates to a two- or three-fold level of concern with security compared to all other concerns. The message for SDN and NFV suppliers and open systems/source developers is clear: Prioritize security capabilities above all with respect to new developments.

While security is top of mind for operators, network service levels are also a major concern. The strong showing for network service levels (59 percent of respondents designate this as a primary concern) certainly points to the increasing awareness that the world’s networks are not keeping pace with demand. Big data, video, complex transactions, and demanding users are all exposing the shortcomings of our networks. And as they do, service levels are declining. That puts more pressure on systems and staff just to keep the network running, leaving less and less time for network improvements that could possibly better serve the end user and, more importantly, the business.

The high concern for networking costs also indicates that our networks are being “fixed” more than enhanced. Unplanned patches always cost more, both short-term and long-term.

Later survey results certainly shine a spotlight on operator hopes for SDN, NFV, open systems, and open source not only to improve network security and service levels but also to drive cost savings across both systems and staff. 

Driving change: operating now over optimizing new

The figure above illustrates the two distinct focal points for the operator. The top five concerns all focus on improving the current state of the network, driving improvements in security, service levels, cost savings, and connectivity. The lesser concerns mostly focus on enhancing the future state of the network, driving new technology adoption, service rollouts, and new user/service provisioning.

We easily could argue that network complexity is at the root of most, if not all, of the highlighted networking concerns, but “overall network complexity” appeared second to last on this list. Complexity opens the door for hackers, undercuts service levels, drives increased costs, and limits new technology/service/user deployments. The message is clear: Make the network simpler, and it is more effective and efficient. Simplicity serves to improve the network’s “now” state as well as its “new” state.

3 SDN and NFV: plans, promises, and problems

SDN and NFV represent a sea change in networking. Together and separately, both target cost savings, operational complexity, and network optimization – and both hold much promise for the operator. Unfortunately, as with all things offering great potential rewards, many possible risks are associated with SDN and NFV rollouts.

Key findings: expected and unexpected

Given the industry attention (and hype) focused on SDN and NFV, any survey is challenged to expose new plans, promises, or problems with these two sea-changing technologies, but this survey took on the challenge – with surprising results across several fronts.

  • The survey revealed that the majority of the 600 survey respondents will have deployed SDN/NFV solutions by the end of 2015. While most respondents may not meet this goal, these aggressive timelines reflect high hopes for SDN and NFV solutions – and also closely related open systems and open source technologies.
  • Confusion surrounding costs, capabilities, complexity, and conversion could inhibit SDN and NFV adoption. Complete answers to all the many questions surrounding real-world SDN and NFV deployments – and subsequent returns – are still lacking. The longer the questions go unanswered, the more operator timelines must be pushed out. Survey responses outlined later in this report certainly indicate that open systems and open source offer some clarity for operators looking for answers to the many questions surrounding SDN and NFV.
  • SDN and NFV benefits must drive high impact immediately. The “here and now” are far more important than the “there and then.” If SDN and NFV are to prove their worth from the start – and justify further advancement by suppliers and acceptance by operators – significant improvements must be seen immediately in such critical areas as security, resource utilization, operational simplicity, and cost savings.
  • The data center has been the primary target for most SDN and NFV developments and solutions to date. For service providers, this is the right place for investment, but for enterprises, SDN developments may be missing the mark. The WAN is the No. 1 target for initial SDN deployments in the enterprise. But for service providers, the LAN is the No. 2 target, and wireless WAN is No. 3. Imagine WAN (core and access) and MAN being last on the list for service providers. For enterprises, behind the WAN and data center, wireless LAN is the No. 3 target – not the campus LAN (core or access), the branch WAN, or the MAN.

Deployment timeline: hope or reality?

SDN and NFV will see widespread deployment by the end of 2015. This may sound aggressive, but real-world challenges (e.g., migration costs, multivendor interoperability, mismatched capabilities, and immature solutions) facing both suppliers and operators will surely push this timeline out for all.

However, the good news is that with such aggressive timelines planned by operators, high hopes will surely drive heightened activity in solutions development, validation, and availability. (As indicated later in this report, operators seem to believe that open systems and open source are key to lowering costs, improving interoperability, providing more consistent capabilities, and solidifying available solutions.) While heightened activity is not likely to deliver widespread SDN and NFV deployments in the next 12 to 24 months, it will certainly accelerate adoption in that key two-to-five-year timeframe. Given that major technology shifts in networking typically take between 10 and 20 years to gain mainstream acceptance, five years seems fast and furious.

Figure 2

Figure 3

Moving to SDN: the interesting catalysts and contradictions

As in the “State of the Network” section, the “operating now” concerns drive SDN and NFV interest and investment more than the often-hyped “optimizing new” capabilities do. This same “here and now” focus comes through loud and clear as respondents cite the expected benefits of SDN and NFV. Making the network more secure, efficient, simpler, and lower-cost form the top five benefits.

Interestingly, advancements in network provisioning and programmability – two well-promoted SDN and NFV benefits – are not cited as top-five benefits. However, these two “new” networking gains are not far behind the more mundane “now” benefits. While the “new” may not drive initial success of SDN and NFV, provisioning and programmability are poised to have a big impact on future networks. 

Figure 4. The benefits of SDN/NFV

The view on cost: capital spending versus operational savings

Reducing or at least reining in networking costs is a high priority for operators and a strong influence over operator SDN and NFV rollouts. Interestingly, all operators give almost equal consideration to capital and operating expenses when evaluating their network state and possible network solutions. While this equalization may fly in the face of network budget breakdowns, which typically show around 75 percent for opex and 25 percent for capex, SDN and NFV solutions must deliver cost savings on both fronts if the promises of SDN and NFV are to be fully realized.

In some ways, that “now” over “new” focus of SDN and NFV is also at work here. Equipment spending, while only around 25 percent of the overall network budget, comes in waves – often big waves. Capex remains a very visible concern and consideration for operators. Capex savings can mean just as much as savings on the operations side of the fence. While the absolute savings will always be less on the capex side, SDN and NFV solutions that reduce capex – or at least avoid making big capex waves – will continue attracting strong operator interest despite the many math exercises that attempt to dismiss capex savings (or spending) as inconsequential. The survey results bear this out.

Barriers to adoption: fighting on many fronts

With any emerging technology, a range of barriers works to prevent adoption. Certainly the advantages offered by onrushing open systems and open source technologies give possible answers to many of these current barriers. The survey results outlined in the following open source section point to an operator base hoping for “open” answers to many SDN and NFV questions they have now.

Figure 5

SDN in the network: not the obvious places and priorities

This is the one area of the survey with clear differences between enterprise respondents and service providers.

For the enterprise, the WAN (31 percent), data center (26 percent), and WLAN (17 percent) are the highest priority targets for SDN deployments. The WAN in the No. 1 slot should be a surprise to most. After all, the data center has been getting the large bulk of the attention in SDN developments and deployments. But on reflection, the WAN makes perfect sense. It is an area of high cost and complexity. The WAN is also a security nightmare. Across the WAN, maximum access meets minimal control. While the data center is a big area of spending and a critical resource for enterprises, it is relatively contained and has been fed a constant stream of advanced technologies for the last decade or so (e.g., virtualization, fast and flexible fabrics, high-performance security systems, massive switches and bandwidth, and automated management). No wonder enterprises want to see SDN applied first to the WAN. The WLAN emphasis also makes sense given the rising complexity and potential risks of the increasingly common BYOD movement. Here SDN can be seen as a way to finally fully integrate the wireless worlds and the wired worlds.

For service providers, the data center is the consensus No. 1 target for SDN and NFV deployments. Given all the activity focused on cloud computing, content delivery, and anything-as-a-service (XaaS) offerings, the service provider data centers must advance across many fronts (e.g., security, automation, analytics, and provisioning) to be successful. What is surprising is the selection of the LAN (16 percent) and wireless WAN (14 percent) as the No. 2 and No. 3 targets for the service provider SDN deployments. WAN oversubscription – and related overspending – would seem a natural No. 2 target for SDN deployments, and yet the WAN (core and access) and the MAN are given short shrift in service provider SDN and NFV plans. 

Figure 6

Figure 7

SDN and NFV: moving in lockstep for service providers

SDN and NFV are separate technology movements driven by different standards groups, mostly distinct technologies, and divergent solution sets, and while both bring software to the forefront in networking, they represent very different migration challenges for service providers. SDN migration requires major changes within the core of provider networks, while NFV solutions can be deployed as incremental add-ons or replacements for existing specialized systems, and yet in the minds (and plans) of service provider operators, SDN and NFV move in lockstep within their network infrastructures. Deployment timelines for both SDN and NFV are almost identical for service providers.

Is it the software-centric nature of both technologies? Is it the anticipated close coupling of SDN’s dynamics and NFV’s demands? Is it the increasing linkage of networking and computing that drives both forward equally? Is it the ownership of both SDN and NFV by network architects or devops analysts? Whatever the driving force or forces, SDN and NFV are tightly linked.

Figure 8

4 The role of open source in SDN and NFV

SDN got its start with a research project aimed at opening up the network to a new model based on centralized control and distributed forwarding. At its core was a technology called OpenFlow. While OpenFlow served as the initial catalyst for the SDN movement, many existing and developing technologies further SDN advancements. Open source is logically positioned to play a strong role in the future acceptance and ultimate success of SDN and closely related NFV developments and deployments by creating de facto standards through common code development.

Key findings: expected and unexpected

While open systems are naturally perceived as key to SDN and NFV, the results of this survey point to an even tighter linkage than expected. Furthermore, the indicated increased focus on open source technology as a principal delivery vehicle for open SDN and NFV solutions points to an operator population that wants to do networking better this time around.

For these operators, “better” networking seems characterized by greater choice in technology and suppliers, heightened open systems functionality, lower equipment and operating costs, and a network infrastructure that is simpler to operate, more efficient to run, and more ready to adapt to new demands – the next second or the next year.

The survey results support this emphasis on openness by showing the strong bias toward open systems as the preferred technology base for networking solutions; 83 percent of respondents demand or prefer the use of open systems within their networks. However, these results also point out that open systems should not only deliver freedom of choice and lower costs but also deliver proven value if they are to be successful in displacing proprietary networking solutions. Operators will no longer be satisfied by open systems that provide for least-common-denominator functionality.

Figure 9

Survey results further raise the ante for open systems by asserting the importance of open source in delivering SDN and NFV solutions; 95 percent look at open source as a positive attribute of any SDN or NFV solutions. Open source has indeed become mainstream, representing what IT users have come to expect from open source software solutions over the last decade: maturity, robustness, and reliability. Gone are the days of outlaw programming, midnight downloads, and hidden operation.

Figure 10

The following survey responses highlight the increasing influence of open systems and open source in the SDN and NFV movements:

  • Open systems and open source are tightly linked for all operators – enterprise and service providers. Open source technology has clearly developed into the preferred delivery method for open systems. This may be driven by lack of faith in single-vendor solutions or by the success of such open source technologies as Linux and OpenStack, or it might be a sign that operators want greater control over the destiny of their networks – and networked businesses – in the future.
  • Operators, while citing vendor freedom as the No. 1 reason to use open source in their networks, feel strongly that commercial vendors should be the primary source for open source. While sounding a bit oxymoronic, the mix of open source and proven development, testing, delivery, and support creates a powerful combination in networking.
  • Network management is cited as a clear and present opportunity for open source to make a real difference not only across SDN and NFV but also across networking overall. For too long, management has been underserved by network advancements and vendor developments. Operators indicate high hopes across the full spectrum of management, from security to automation to analytics to optimization.
  • Despite all the hope for open systems and open source, a need to overcome perceived challenges still exists, especially when those perceptions closely parallel operator concerns with SDN and NFV as well as networking overall. Misperceptions relating to open source across such critical areas as security and reliability must be dispatched. The good news is the indications that these perceptions are breaking down. Over the last few years, many open source technologies have proven themselves to be fast, reliable, secure, highly functional, simple to deploy and operate, and cost-effective.

The tight linkage between open systems and open source

Operators view open systems and open source as equally important to the success of SDN developments and deployments. A full 76 percent of respondents view open systems as critically (18.3 percent) or very (57.8 percent) important to SDN and NFV. An almost matching 68.5 percent of respondents view open source as critically (13.5 percent) or very (55 percent) important to SDN and NFV. The message is clear: Open source serves as the principal delivery mechanism for SDN and NFV solutions based on industry standards and open systems.

Figure 11

The commercial imperative: the sourcing of open source

To steal an old advertising phrase, “This is not your father’s open source.” Today’s operators demand more from their open source and open source suppliers. Given that network expectations are extremely high, so too are expectations for open source working within our networks. Operators demand that open source be:

  • Created using formal design and development practices
  • Validated by strict integration and testing procedures
  • Delivered and deployed via proven tools and techniques
  • Supported by trusted programs, processes, and people

The vast majority of operators (76 percent) prefer to acquire their open source from a commercial supplier, so they are provided the peace of mind that comes with combining the inherent cost-effectiveness and interoperability of open source with the proven integration, validation, and support practices offered by commercial suppliers. Additionally, operators benefit from the increased supplier focus on networking capabilities that serve to complement open source functions. 

Figure 12

The driving forces for open systems: freedom and function

As expected, avoiding vendor lock-in and the high costs that are all too often associated with lock-in are the top two drivers behind the use of open source in networks. Years ago, the story would have stopped there, but times have changed for open source because operators are demanding more from it. Survey respondents clearly indicate that improved functionality follows closely on the heels of freedom of choice and cost savings. Heightened software capabilities, ease of management, and more frequent enhancements are all significant drivers of open source usage.

Figure 13

Network management: still crazy after all these years

While certain high-profile open source projects garner much of the industry’s attention – e.g., OpenStack and kernel-based virtual machine (KVM) – the mundane occupies the attention of the operator. Respondents strongly favor the application of open source to critical SDN and NFV management functions. The top five applications for open source are all management functions. The top 10 is then rounded out by the more visible SDN- and NFV-related open source efforts focused on hypervisor, controller, orchestrator, and API developments.

Figure 14

Open source: matching real needs to perceived challenges

Within the SDN and NFV movements, open source must overcome significant challenges – both real and perceived – before it fulfills on its promise of freedom, savings, and functionality. An examination of survey results reveals that, for open source, a close match exists between current major concerns with current networks and operator-cited open source challenges.

Figure 15

For example, security vulnerabilities are a major concern with current networks, a potential barrier to SDN and NFV adoption, and a perceived challenge for open source. Open source solutions must be proven to strengthen the security posture of the network. This survey indicates that security concerns are the No. 1 challenge for open source. Is this a real or perceived challenge? A case could be made that open source can be more secure than a single-vendor proprietary system. Despite the best efforts of a single vendor, proprietary solutions are run through a singular design, development, testing, and deployment process. Open source runs through multiple iterations of all stages of development and deployment — the more iterations, the more intense the scrutiny, the more solid the resulting solution. The onus is on the open source developers to see that this multi-pronged approach serves to bolster not only security but also other cited challenges, such as reliability, performance, interoperability, and operational simplicity.

5 Appendix A: SDN/NFV survey

This survey targeted two main areas of interest in SDN and NFV. First, responses serve to highlight the drivers, barriers, timelines, and targets of SDN and NFV solutions for both enterprises and service providers. Second, responses serve to outline the expected role that open systems and open source will play in the advancement and adoption of SDN and NFV solutions.

Survey methodology and definitions

In December 2013, 600 respondents completed an online Gigaom Research survey focused on SDN, NFV, open systems, and open source within medium-to-large organizations (500 employees served as the low end of our employee count). Of the 600 respondents, 300 represented enterprises, and 300 represented service providers.

Two distinct, though similar, surveys solicited responses from the two groups. Each included unique questions for each set of respondents. For example, service providers were asked to provide service and customer profile information as well as specific input on NFV – a technology driven mainly by service providers – while enterprises were asked to provide information on industry affiliation and internal locations served. (See Appendix B for a copy of the enterprise and service provider versions of the survey.)

For clarity, we use the following definitions to describe the main focal points of this survey and report:

  • Software-defined networking (SDN) – An approach to networking that is characterized by a decoupling of control and forwarding functions, enabling simplified operations, heightened automation, improved resource efficiency, and on-demand network programmability.
  • Network functions virtualization (NFV) – The deployment and delivery of networking services via software systems executing as server-based processes. These server-based software systems serve as direct replacements for traditional physical networking appliances (e.g., WAN optimization) and devices (e.g., routers).
  • Open systems – Systems (e.g., components, protocols, interfaces) that conform to either formally adopted industry standards or de facto published specifications. Open systems are independent of any single particular supplier or exclusive group of suppliers.
  • Open source – Software source code that can be acquired, used, redistributed, and modified free of charge by any individual or organization.

Survey respondents

A total of 600 medium and large organizations provided responses for this survey effort: 300 enterprises and 300 service providers. All respondents were asked for a job title. Survey respondents represented a broad mix, from IT executives (e.g., CIO, vice president, directors) to overall area IT managers (e.g., senior managers, managers) to specialized IT technologists (e.g., architects, engineers, analysts, and even a few developers). 

Figure 16

The enterprise respondent

The 300 enterprise respondents represented a broad mix of vertical industries. Key industries such as financial, manufacturing, and healthcare were well represented. 

Figure 17

In terms of size of the enterprises included in the responses, the following applies:

  • 47 percent of responding enterprises have more than 2,500 employees
  • 26 percent have from 1,001 to 2,500 employees
  • 27 percent have from 500 to 1,000 employees

This bias toward larger enterprises is also reflected in the overall complexity of respondent networks. Results indicate that 47 percent operate networks with more than 20 sites and another 33 percent operate networks with six to twenty sites. These numbers indicate that even though half of the respondents operate within enterprises with fewer than 2,500 employees, their networks are sizeable and well distributed. 

Figure 18

The service-provider respondent

The service-provider respondents also represent a diverse set of primary businesses. The results provide a view into the changing nature of the service provider. Almost half the respondents offered cloud-computing services. More than a third offered traditional telecom services. Hosting, content delivery, and wireless services were also well represented as parts of the services mix.

Figure 19

Of the service providers included in the responses:

  • 61 percent have more than 2,500 employees
  • 21 percent have from 1,001 to 2,500 employees
  • 18 percent have from 500 to 1,000 employees

This bias toward larger service providers is also reflected in the number of customers served and the number of customer locations served. Results indicate that 69 percent of the respondents service more than 2,000 customers and that these customers are sizeable.

Looking at customer locations served by responding service providers:

  • 41 percent serve more than 10,000 customer locations
  • 17 percent serve 5,001 to 10,000 locations
  • 18 percent serve 1,001 to 5,000 locations
  • 24 percent serve 100 to 1,000 locations

Figure 20

6 Appendix B: the survey instruments

Enterprise survey instrument

Examining the Role of Open Source Technology in SDN

THE ENTERPRISE VIEW

Introduction

As our reliance on network connectivity increases by the day, so does the complexity involved in deploying and managing those networks. This challenge has given rise to a new era of networking defined by software-defined networks (SDN). These SDNs are more flexible, automated, efficient, and responsive to business and user requirements.

Demographic Questions

Industry affiliation? (Retail, Health Care, Government, Manufacturing, Financial, Other______)

Size of organization by number of employees? (100-500, 501-1000, 1001-2000, >2000)

Number of locations served by your network? (1, 2-5, 6-20, >20)

Status of SDN within your organization? (Researching, Actively planning, Evaluating, solutions, Piloting, Point deployment, Wide-scale deployment)

Current State of The Network

On a scale of 1 to 10, rank your network’s ability to serve your organization’s needs (with 1 being poorest possible service and 10 being greatest possible service). ________

Rate 1 to 5, your level of concern with your current network in the following areas (1 being of little concern and 5 being of greatest concern):

___Cost of networking equipment

___Cost of network operations

___Network connectivity – available/affordable bandwidth, wireless access, server/storage I/O

___Network service levels – uptime, response times, latency, accessibility, etc.

___Overall network complexity

___Security vulnerabilities

___Slow to adapt to new user and business requirements

___Ability to leverage new technology in a timely fashion

___Failure to derive maximum business value from the network

What is the current level of end user satisfaction with your network?

___Completely satisfied

___Mostly satisfied

___Somewhat Satisfied

___Mostly not satisfied

___Not Satisfied at All

What is the current level of business management satisfaction with your network?

___Completely satisfied

___Mostly satisfied

___Somewhat Satisfied

___Mostly not satisfied

___Not Satisfied at All

When evaluating networking solutions, which one of the following statements best represents your level of consideration of open systems that support industry standards?

___Demand the use of open systems wherever possible.

___Prefer the use of open systems, but will accept proprietary substitutes that add proven value.

___Favor proprietary solutions with strong potential to add greater value than open systems.

SDN Benefits and Challenges

Rate 1 to 5, the importance of SDN benefits to your organization (1 being of little importance and 5 being of critical importance):

___Lower network capital expense (CAPEX)

___Lower network operating expense (OPEX)

___Simplify network deployments and operations

___Strengthen the security posture of your network

___Heighten utilization of networking (e.g., bandwidth) and networked (e.g., servers) resources

___Enable on-demand network programmability

___Accelerate the provisioning of new sites, users, applications…

Where in your network will SDN have the biggest impact initially? (Select one.)

___Data center network

___Wide area network (WAN) Core

___Branch WAN

___Campus LAN – Core

___Campus LAN – Access

___Wireless LAN

___Metro Area Network (MAN)

What is your level of satisfaction with the current vendor offerings for SDN?

___Completely satisfied

___Mostly satisfied

___Somewhat Satisfied

___Mostly not satisfied

___Not Satisfied at All

What is your biggest concern with current SDN solutions? (Select one.)

___Expense of investing in new hardware platforms and software systems

___Lack of interoperability of SDN/NFV solutions

___Lack of clarity and consistency of features/functions across SDN solutions

___Lack of proven performance (e.g., throughput, availability…) of SDN solutions

___Other (Please specify.) ______________________________________

Rate 1 to 5, the following barriers to SDN adoption (1 representing a barrier of little consequence and 5 representing a significant barrier):

___No technical or business requirements driving an SDN rollout

___Limited availability and unproven status of SDN technologies and solutions

___Shortcomings of industry standards and open systems supporting SDN

___Difficulty in migrating our existing network infrastructure to SDN

___SDN increases operational complexity and integration/test burden during deployment

___Risks (e.g., failures, slowdowns) outweigh potential rewards (e.g., savings/services boost)

___Advanced SDN capabilities (e.g., on-demand network programmability) not needed

SDN and Open Source Technology

What is your view of the role of industry standards and open systems in the development and deployment of SDN solutions?

___Critically important

___Very Important

___Somewhat important

___Of little importance

___Not important at all

What is your view of the role of open source technology in the development and deployment of SDN solutions?

___Critically important

___Very Important

___Somewhat important

___Of little importance

___Not important at all

What are the biggest advantages of applying open source technology to SDN solutions and deployments? (Select three.)

___Avoid vendor lock-in.

___Lower acquisition and maintenance costs.

___Gain access to large library of networking software systems.

___Boost networking software capabilities.

___Accelerate networking software enhancements.

___Assure software systems interoperability.

___Ease management of network — and networking software.

___Leverage top developer talent and software best practices.

___Other (Please specify.) ______________________________________

What are the biggest drawbacks of applying open source technology to SDN solutions and deployments? (Select three.)

___Lack of commercial vendor support

___Incompatible with existing network

___Performance shortcomings

___Feature shortcomings

___Reliability concerns

___Security concerns

___Deployment challenges – e.g., increased integration and test burden

___Operational complexity – e.g., non-standard management interfaces and tools

___Legal concerns – e.g., license management and compliance

___Other (Please specify.) ______________________________________

Rate 1 to 5, your intention to apply open source technology to the following SDN functions (1 being least likely to apply and 5 being most likely to apply).

___SDN controller

___Network automation

___Network monitoring and analytics

___Northbound APIs

___Southbound APIs

___Policy management

___Security services – e.g., firewall, intrusion prevention…

___Network optimization – e.g., application delivery controller, WAN optimization…

___Orchestrators – e.g., OpenStack and associated plug-ins

___Hypervisor – e.g, Kernel-based Virtual Machine (KVM)

___Other (Please specify.) ______________________________________

What open source technology projects have you contributed to or are currently using?

[Accept as many as cited.]

 

Service provider survey instrument

Examining the Role of Open Source Technology in SDN/NFV

THE SERVICE PROVIDER VIEW

Introduction

As our reliance on network connectivity increases by the day, so does the complexity involved in deploying and managing those networks. This challenge has given rise to a new era of networking defined by software-defined networks (SDN) and new network service deployment models based on network functions virtualization (NFV). These software-centric networks and functions are more flexible, automated, efficient, and responsive to business and user requirements.

Demographic Questions

Service focus? (Telecom services, Wireless services, Hosting services, Content delivery services, Cloud Computing services, Mix of the above, All of the above)

Size of organization by number of employees? (100-500, 501-1000, 1001-2000, >2000)

Number of customers served by your organization? (100-500, 501-1000,1001-2000, >2000)

Number of customer locations served by your organization? (100-1000, 1001-5000, 5001-10000, >10000)

Status of SDN/NFV within your organization? (Researching, Actively planning, Evaluating solutions, Piloting, Point deployment, Wide-scale deployment)

Current State of The Network

On a scale of 1 to 10, rank your network’s ability to serve your organization’s needs (with 1 being poorest possible service and 10 being greatest possible service). ________

Rate 1 to 5, your level of concern with your current network in the following areas (1 being of little concern and 5 being of greatest concern):

___Cost of networking equipment

___Cost of network operations

___Network connectivity – available/affordable bandwidth, wireless access, server/storage I/O

___Network service levels – e.g., uptime, response times, latency, accessibility, etc.

___Overall network complexity

___Security vulnerabilities

___Ability to leverage new technology in a timely fashion

___Delays in implementing and introducing new service offerings

___Slow to provision new customers and new services for existing customers

What is the current level of external (e.g., customer) satisfaction with your network?

___Completely satisfied

___Mostly satisfied

___Somewhat Satisfied

___Mostly not satisfied

___Not Satisfied at All

What is the current level of internal (e.g., business mgmt) satisfaction with your network?

___Completely satisfied

___Mostly satisfied

___Somewhat Satisfied

___Mostly not satisfied

___Not Satisfied at All

When evaluating networking solutions, which one of the following statements best represents your level of consideration of open systems that support industry standards?

___Demand the use of open systems wherever possible.

___Prefer the use of open systems, but will accept proprietary substitutes that add proven value.

___Favor proprietary solutions with strong potential to add greater value than open systems.

SDN/NFV Benefits and Challenges

Rate 1 to 5, the importance of SDN/NFV benefits to your organization (1 being of little importance and 5 being of critical importance):

___Lower network capital expense (CAPEX)

___Lower network operating expense (OPEX)

___Simplify network deployments and operations

___Strengthen the security posture of your network and customer networks

___Heighten utilization of networking (e.g., bandwidth) and networked (e.g., servers) resources

___Enable on-demand network programmability

___Accelerate the provisioning of new customers and services

Where in your network will SDN/NFV have the biggest impact initially? (Select one.)

___Data center

___LAN Core

___WAN Core

___WAN Access (Last Mile)

___Wireless WAN

___Metro Area Network (MAN)

What is your level of satisfaction with the current vendor offerings for SDN/NFV?

___Completely satisfied

___Mostly satisfied

___Somewhat Satisfied

___Mostly not satisfied

___Not Satisfied at All

What is your biggest concern with current SDN/NFV solutions? (Select one.)

___Expense of investing in new hardware platforms and software systems

___Lack of interoperability of SDN/NFV solutions

___Lack of clarity and consistency of features/functions across SDN/NFV solutions

___Lack of proven performance (e.g., throughput, availability…) of SDN/NFV solutions

___Other (Please specify.) ______________________________________

Rate 1 to 5, the following barriers to SDN/NFV adoption (1 representing a barrier of little consequence and 5 representing a significant barrier):

___No technical or business requirements driving an SDN/NFV rollout

___Limited availability and unproven status of SDN/NFV technologies and solutions

___Shortcomings of industry standards and open systems supporting SDN/NFV

___Difficulty in migrating our existing network infrastructure to SDN/NFV

___Risks (e.g., failures, slowdowns) outweigh potential rewards (e.g., savings/services boost)

___Advanced SDN capabilities (e.g., on-demand network programmability) not needed

___Limited performance and scalability of NFV solutions

SDN/NFV and Open Source Technology

What is your view of the role of industry standards and open systems in the development and deployment of SDN/NFV solutions?

___Critically important

___Very Important

___Somewhat important

___Of little importance

___Not important at all

What is your view of the role of open source technology in the development and deployment of SDN/NFV solutions?

___Critically important

___Very Important

___Somewhat important

___Of little importance

___Not important at all

What are the biggest advantages of applying open source technology to SDN/NFV solutions and deployments? (Select three.)

___Avoid vendor lock-in.

___Lower acquisition and maintenance costs.

___Gain access to large library of networking software systems.

___Boost networking software capabilities.

___Accelerate networking software enhancements.

___Assure software systems interoperability.

___Ease management of network — and networking software.

___Leverage top developer talent and software best practices.

___Other (Please specify.) ______________________________________

What are the biggest drawbacks of applying open source technology to SDN/NFV solutions and deployments? (Select three.)

___Lack of commercial vendor support

___Incompatible with existing network

___Performance shortcomings

___Feature shortcomings

___Reliability concerns

___Security concerns

___Deployment challenges – e.g., increased integration and test burden

___Operational complexity – e.g., non-standard management interfaces and tools

___Legal concerns – e.g., license management and compliance

___Other (Please specify.) ______________________________________

Rate 1 to 5, your intention to apply open source technology to the following SDN/NFV functions (1 being least likely to apply and 5 being most likely to apply).

___SDN controller

___Network automation

___Network monitoring and analytics

___Northbound APIs

___Southbound APIs

___Policy management

___Security services – e.g., firewall, intrusion prevention…

___Network optimization – e.g., application delivery controller, WAN optimization…

___Orchestrator – e.g., OpenStack and associated plug-ins

___Hypervisor – e.g., Kernel-based Virtual Machine (KVM)

___Other (Please specify.) ______________________________________

What open source technology projects have you contributed to or are currently using?

[Accept as many as cited.]

7 About Mark Leary

In 2012, 30-year IT industry veteran Mark Leary founded The First Tracks as a research, analysis, and consulting firm focused on disruptive network technologies, leading-edge solutions, and groundbreaking operator deployments. Prior to leading The First Tracks, Leary worked for Cisco Systems for 12 years, during which time he formulated and executed market and business development activities relating to Cisco’s core technology systems (routers, switches, and IOS), application networking services, and sustainable green IT solutions. Before his work at Cisco, Leary was vice president of networking industry research and consulting at IDC (International Data Corporation). During his 13 years at IDC, he functioned as an industry analyst and IT consultant to Fortune 500 companies, networking and computing system suppliers, software vendors, service providers, and the investment community. Some of his most prominent accomplishments in the industry include driving adoption of new technologies and best practices within Fortune 5000 networks; authoring technology and business articles for such publications as the New York Times, Computerworld, and Network World; and evangelizing “The Network” across the world—from private executive briefings to public seminars, panels, and conference sessions. Leary received a bachelor of science degree in computer science as well as an MBA in business strategy and planning from Boston College. 

8 About OpenDaylight Project

The OpenDaylight Project is a collaborative open source project that aims to accelerate adoption of software-defined networking (SDN) and create a solid foundation for network functions virtualization (NFV) for a more transparent approach that fosters new innovation and reduces risk. Founded by industry leaders and open to all, the OpenDaylight community is developing a common, open SDN framework consisting of code and blueprints. Get involved at www.opendaylight.org.

OpenDaylight is a collaborative project at The Linux Foundation. Linux Foundation Collaborative Projects are independently funded software projects that harness the power of collaborative development to fuel innovation across industries and ecosystems. Find out more at www.linuxfoundation.org.

9 About Gigaom Research

Gigaom Research gives you insider access to expert industry insights on emerging markets. Focused on delivering highly relevant and timely research to the people who need it most, our analysis, reports, and original research come from the most respected voices in the industry. Whether you’re beginning to learn about a new market or are an industry insider, Gigaom Research addresses the need for relevant, illuminating insights into the industry’s most dynamic markets. 

Visit us at: research.gigaom.com.

10 Copyright

© Knowingly, Inc. 2014. "SDN, NFV, and open source: the operator’s view" is a trademark of Knowingly, Inc. For permission to reproduce this report, please contact sales@gigaom.com.

Tags