I’ve heard from several telecom managers who are assessing MPLS-based services for their WANs. Given the growing momentum behind the technology, it makes sense to ask whether MPLS-based services are right for your organization.
I’ve heard from several telecom managers who are assessing Multi-protocol Label Switching-based services for their WANs. Given the growing momentum behind the technology, it makes sense to ask whether MPLS-based services are right for your organization.
First off, MPLS is a technology, not a service. Most carriers run MPLS underneath a wide range of services, including frame relay, wide-area Ethernet, native IP and ATM. The advantages accrue primarily to the carrier. User benefits include lower cost in most cases, greater control over networks, and more detailed QoS. In fact, QoS is the primary reason IT executives opt for MPLS – in a recent Nemertes benchmark, 62% of organizations told us they’re using MPLS today or plan to deploy it, with 55% listing QoS as the main reason.
MPLS-based services are a good fit in the following scenarios:
Your company has a lot of any-to-any traffic. Any-to-any traffic requires N-squared number of connections – an expensive proposition in network technologies that charge by the circuit, such as frame or ATM. Most companies don’t have a lot of any-to-any traffic, unless they’re engaged in a convergence project. The majority of today’s applications tend to be client/server, which generate hub-and-spoke traffic patterns. For these, switching to MPLS doesn’t buy much: Firms report around 10% cost savings as compared with legacy frame or ATM. But the scenario changes dramatically when MPLS is used to converge voice and video – or with next-generation software architectures.
You’re planning a convergence project. Most firms see immediate savings – 25% or more – when they begin combining voice and video traffic over the MPLS WAN. Video often is carried over ISDN circuits that are expensive. Consolidating this traffic onto a data network can eliminate the need for an ISDN network, generating immediate savings. Also, both video and voice tend to have any-to-any traffic patterns, unlike legacy data apps – so the any-to-any cost savings begin to kick in.
You’re planning to deploy next-generation computing infrastructure such as Web services, peer-to-peer or grid computing. Web services and peer-to-peer generate any-to-any traffic patterns; grid computing does the same, and often requires QoS capability. In fact, for some financial services firms, grid computing is the primary driver behind MPLS.
If any of these scenarios apply, you’ll want to look into MPLS-based services. But please don’t jump on the bandwagon just because MPLS is “the next big thing” – you should sort out what MPLS can and can’t do for you before taking the plunge.
Want to learn more? Check out MPLSCon in New York this week. You’ll see real-world examples of MPLS deployments from the U.S. Department of Agriculture, the state of Illinois and others. (Full disclosure: Both yours truly and my co-columnist Scott Bradner will be keynoting.)
