When we think about application service providers and building out ecosystems, we ask questions such as: How do enterprises want to consume the application? How do they want to consume network? Are they sitting in one location or in multiple facilities? How do you know the network can be extended to where the user is, wherever they are in the world? How do you think in terms of the physical network route as well as the logical distribution of the network across the backbone and access networks? And how do you provide solution transparency?
When you look at what is going on in our world, the application services ecosystem continues to be built out with elements including transport networks; IP networks; data centers; enterprises; and the application providers - network complexity is everywhere. This complexity increases as the number of different elements and the type of interactions with the network increase. In that context, the user is constantly looking for simplicity. While we are being burdened with more complexity as service providers, the user expects and demands a reliable and simplified experience. So it’s incumbent upon us, as we build out the ecosystem of an interconnected application system, to understand how to simplify that experience and eliminate the complexities.
The Network as a Key Defining Element of the User Experience
Looking at the ecosystem from an application perspective, we see computing power and usage increasingly tied to a network-centric view. And that network-centric environment is driven by the user experience. The enterprise customer wants to distribute their applications into a hosted environment to get the economic and performance benefits of the third-party data centers, but the user experience must be maintained or improved over time - and increasingly, the network is a key defining element of that user experience. That places more value on the network; both in terms of economic reality of what enterprise customers are willing to invest to get that user experience, as well as the value in service availability.
In designing an application connectivity solution, one of the things we focus on is how customers are consuming the data in addition to understanding the physical networks and how to connect back to the application. A consideration that illustrates this example is the globalization of information - not only for where the users of those applications sit, but how they interact with their ecosystem of vendors and customers. So physically interconnecting networks together and understanding how to execute around them is one of the key aspects of where that marketplace of networks and that ecosystem need to be.
Key Network Attributes for an Application Provider
Applications, whether they are virtualized cloud-based applications, messaging, live collaboration or basic web-based services, consume more and more bandwidth with the increase of application usage and complexity. The amount of available bandwidth and the reliance on the network and its service capabilities are relevant to the successful delivery of application services. Relevant in terms of customers wanting to define specific Service Level Agreements (SLAs) for application availability; driving the service provider to understand what that network looks like on Layers 4 through 7, not just the transport layers, and vice versa for the application provider. When we say performance is critical, performance can mean different things to different types of customers. In general, what we are looking at is more and more enterprises are placing performance attributes around availability, latency, and jitter. This means understanding how the application is interacting with the network to ensure that not only the primary routes access the applications, but also the secondary and failover routes maintain a process that keeps the user experience live, especially for real-time applications such as video and collaboration tools.
Building an Ecosystem for Application Providers
How do you design an efficient application network that deals with increased complexity yet delivers a positive user experience? Additionally, you need to consider the optimal way for the ecosystem connectivity to be designed from application to users, who are typically in a multi-location global environment with mobility requirements. Network design options rely on the understanding of the physical, logical routes that interact with the application. In today’s world, no one owns network everywhere. The largest suppliers in the world own single-digit percentages, at best, of the world’s infrastructure from a fiber resource perspective. In that context, when you take the U.S. for example, there are 903 different physical LEC owners of infrastructure in the marketplace. How do you interact in this environment considering the economic reality that directly connecting to the infrastructure typically brings the best performance? If there are 16 different networks that are optimal to interconnect the enterprise back to the applications, how do you execute around that?
Global Capacity addresses the physical challenges of interconnecting to multiple networks by aggregating them together at central locations like Equinix facilities around the country. Furthermore, our network hubs simplify the commercial aspect of interconnections with predefined terms, conditions, service delivery and NOC processes while providing users with visibility into network connectivity and performance - key attributes to ensuring business application services are being delivered optimally.
For more information about how Global Capacity’s One Marketplace can improve your application delivery, click here.
Schedule a meeting with Global Capacity to review your application connectivity requirements.
Global Capacity's Chief Revenue Officer Ben Edmond participated at the Light Reading Ethernet & SDN EXPO presenting 'A Service Providers Perspective: Connecting An Ecosystem of Application Providers' which took place on October 3, 2013 during the Workshop: Wholesale Services & Interconnection In The CE 2.0 Era