Extreme’s Momentum and Market Leadership in the IXP/CXP Space

Internet and Cloud Exchange Points (IXP/CXP) provide low cost and highly efficient interconnection and exchange of traffic between multiple independent IP networks.  Effectively, these exchanges provide the ability to share services between public, private and enterprise domains.

We’ve written recently about how operators of these exchange points can become key advisors for their enterprise customers and work with enterprises to determine their optimal cloud strategy.  

Here, we’ll address key challenges faced by the IXP/CXP space as it meets the needs of a growing customer base (which may include network service providers, content providers, and enterprises) and downstream constituencies (which in turn includes enterprises as well as consumers). The business demands on this sector have grown very complex, which has led to new requirements in terms of automation, telemetry, and adaptive, high-density platforms.

What Do Enterprise Customers Expect of IXPs?

When enterprises select exchanges to meet their mission-critical business requirements and budgets, they are increasingly demanding lowered costs and more stringent SLAs. This makes it incumbent on the IXPs/CXPs to automate common tasks to free up their cycles, and offer new troubleshooting tools and value-added services.  

As part of delivering new services, IXPs and CXPs must increasingly offer value-added services; examples include higher bandwidth, tiered bandwidth, higher SLA, improved visibility/reporting, flow optimization, and hybrid cloud services.

As IXPs/CXPs strive to fulfill these requirements, they face a variety of challenges in helping reducing costs and increasing revenue opportunities:  

• Cost/port is a key factor as customer connections and services migrate from 10Gbps to 100Gbps. Ultra-high density 10Gbps/100Gbps platforms are essential with the right mix of technologies (e.g., VPLS and EVPN, etc.)   
• IXPs/CXPs are confronting the inefficiencies of manual tasks or one-off non-reusable and inflexible, partial automation. There are rising operational costs associated with handling increasingly complex infrastructure for growing numbers of customers who need cloud services (private and public).
• Slow service delivery means longer time to revenue, poor customer experience and stalled access to auxiliary revenue streams. This also increases chances for human error and opens the potential for customers going to competitors. 
• Further revenue loss can come from not offering a variety of connection services. For example, IXPs/CXPs could benefit from offering more tiered service options such as a fractional service (e.g., 20G off a 100G link).

IXPs and CXPs are in a price-sensitive business. Without the required technology mix and automation, IXP/CXP OpEx can continue to be very high. Lack of a future-proofed platform and agility could lead to higher TCO, margin erosion, and suboptimal/rigid customer support tools.

How Are Leading IXPs/CXPs Addressing This? 

Executives in these intersecting industries are asking themselves how are they can address its challenges and opportunities:

• Reduce solution costs, both initially and over time; this involves architectural choices (for service agility and variety) and the need for scalable platforms for future proofing – capacity, density, and modularity are factors
• Reduce time to revenue (identify and deliver new opportunities)
• Drive up operational efficiency (resolve issues faster and more permanently)

For example, AMS-IX asserted that moving to SLX 9850 helped AMS-IX reduce complexity, increase scale, and save cost from fewer core sites, and in turn allows them to be more competitive in gaining new customers. 

Henk Steenman, their CTO, noted that, “This development is helping us prepare our infrastructure for the future. This not only supports ongoing growth, but also fulfills the need for more bandwidth. It will help us manage the increasing amount of traffic, deliver more 100G-ports and get things done in a more cost-efficient way. This, in turn, will allow us to reduce prices further.”

More recently, Telehouse, a global data center provider and KDDI Group company, announced that that they had partnered with Extreme to meet strong market demands for 100G ports. Their solution, called the Centillion II Platform is 400G-ready and consists of Extreme’s SLX routers in both the core and the edge, which significantly increases the number of available 100G ports. 

IX.br in Brazil, which serves 110 million users in the fourth largest Internet market in the world,  also deployed the Extreme SLX 9850 and SLX 9540 routing solutions in its Internet Exchange Point in São Paulo to efficiently manage daily network traffic peak of about 3.0 Tbps.  “The network is the most important part of an Internet Exchange, and must be cost-effective, agile, robust and reliable,” said Julio Sirota, manager of IX.br infrastructure.  “We’ve seen a growing requirement to scale 100GE port density of the IX.br platform, and we expect that trend to continue.”

Leveraging Automation

Event-driven automation is another key requirement for Internet Exchanges. NL-ix utilizes StackStorm at the core of its strategy to deliver on the vision of automated software service offerings and business processes (case study here). As Diego Luis Neto, a Senior Software Engineer at NL-ix, stated, “StackStorm enables us to develop complex workflows—for both customer-facing and internal business processes—across all departments in the company.”  

By leveraging Extreme solutions, IXPs/CXPs have a pathway to reduce time to revenue and to turn up and start billing for new services in minutes, hours or days instead of weeks or months. Use of the Extreme Workflow Composer including its automation suites can lead to higher-margin services and free cycles to develop new service and application offerings.

Conclusions 

To obtain maximal CapEx and OpEx efficiency, IXPs/CxPs (and indeed all operators) must optimize: 

• Density: Achieve the lowest cost per bit of data that moves across the network using a highly efficient platform
• Scalability: Obtain a more continuous way of scaling their Infrastucture to meet short-medium term requirements
• Visibility: Increase service uptime and tighten SLAs (without increased costs) to enable greater competitiveness and/or perceived value; also provide the collected data needed for intelligent automation
• Automacity: Leverage intelligent automation to create margin opportunities for services delivered

For a comprehensive discussion on Extreme’s data center platforms, and relevant feature sets to the IXP/CXP market, see our article on solutions for cloud data centers. For further information, follow the links in the article and consult your Extreme representative.  

Finally, Extreme will be attending and presenting at Global Peering Forum in Seattle, WA from April 10-13 and Euro-IX in Galway, Ireland from April 15-18. Come see us there.

This post was originally published by Product Marketing Director Alan Sardella.