Optical/IP Networks

Survey Explodes MPLS VPN Myths

A worldwide survey of more than 400 carrier executives, engineers, and operational staff has put to rest speculation over whether service providers are serious about deploying IP VPNs and Ethernet services over Multiprotocol Label Switching (MPLS) backbones.

In a nutshell, the survey -- the results of which were released today by Heavy Reading, Light Reading's market research division -- demonstrates that carriers are indeed deadly serious (see HR Survey Points to Big VPN Changes).

The rewards of rolling out Layer 3 IP VPN services and Ethernet services outweigh the risks, according to the majority of the survey respondents. The same holds true for carriers planning to offer services analogous to private lines and Frame Relay services over their MPLS backbones.

For an executive summary of Heavy Reading's report, entitled "2004 Survey of Carrier Attitudes Toward IP/MPLS Backbones and VPNs," click on this link.

The survey assesses carrier attitudes to potential problems with MPLS, ranging from questions over quality-of-service guarantees to the absence of standardized operations, administration, and maintenance (OAM) functions (see The Softer Side of Convergence).

It also quantifies concerns over other issues, such as whether enterprise customers will be prepared to hand over control of their router networks to carriers, and whether MPLS-based services will cannibalize lucrative business from legacy services.

Carrier attitudes to the attractions of providing multiple services over converged MPLS backbones are also probed by the survey. The ability to offer customers more scaleable and flexible services, with higher bandwidths, is considered highly important, as is the potential for cutting capital and operating expenditures.

The survey goes on to gauge the way in which carriers balance these pros and cons by asking respondents to predict what proportion of their companies' revenues will come from MPLS-based services this year and next. A significant proportion of carriers expect double-digit percentages by 2005.

One of the conclusions of the survey is that Layer 3 IP VPNs are already being provided by many carriers and have a promising future ahead of them. This flies in the face of assertions by Layer 2 fans, such as the CTO-level employee at one Extreme-ly popular equipment vendor. He says carriers might make plenty of noise about Layer 3 IP VPN services but they don't offer them commercially, partly because the technology's not ready and partly because enterprise users don't want them. This isn't born out by the survey findings.

The survey was taken by a total of 404 people from more than 200 carriers around the world, including representatives from AT&T Corp. (NYSE: T), BellSouth Corp. (NYSE: BLS), British Telecommunications plc (BT) (NYSE: BTY; London: BTA), MCI (Nasdaq: WCOEQ, MCWEQ), NTT Communications Corp., SBC Communications Inc. (NYSE: SBC), Sprint Corp. (NYSE: FON), Verizon Communications Inc. (NYSE: VZ), and many others.

The questions addressed by the survey include the following:

  • Which services do carriers expect to deliver over a converged infrastructure -- legacy offerings like private lines and Frame Relay, or new services such as Ethernet-based products and Layer 3 IP VPNs?

  • How do carriers perceive the potential rewards and risks of providing services over IP/MPLS backbones?

  • What are service providers’ expectations regarding enterprise customer demands for new services? Do they believe enterprise users want Layer 3 IP VPN services, or would most of them prefer to buy Layer 2 Ethernet VPN services and retain control of their own router networks?

  • How widespread is current deployment of Layer 3 VPN technology? How many carriers are actually in position to offer new services now? When do other carriers expect to have needed techonology in place?

  • How concerned are carriers about cannibalizing their legacy services with new offerings like Ethernet services and Layer 3 IP VPNs?

  • When will it become feasible for carriers to jointly provide VPNs? Do carriers even see the need to do so?

  • How much interest do carriers have in service interworking -- providing VPNs that link together sites using different technologies, such as Frame Relay and Ethernet?

  • What are carrier expectations regarding revenues from services running over converged IP/MPLS backbones?

  • What do carriers perceive as the biggest factors that will influence the migration to deploying IP/MPLS-based services?
— Peter Heywood, Founding Editor, Light Reading

Heavy Reading's 43 page report, 2004 Survey of Carrier Attitudes Toward IP/MPLS Backbones and VPNs, costs $2,495. It's accompanied by a searchable online database that allows further statistical analysis by region, carrier type and respondent job function.

Archives of Related Light Reading Webinars:

Sign In