SDN: Enabling the Cloud Backbone of the Future

The Indian cloud market is rapidly maturing. In fact, a recent report titled the India Cloud Market Overview from International Data Corporation (IDC), states that the Indian cloud market grew 70 percent in 2012 and is expected to exhibit a 50 percent growth rate for the next three more years.

The success or failure of public cloud services can be measured by whether they deliver high levels of performance, security and reliability that are on par with or better than those available within enterprise-owned data centers. To meet these requirements, cloud providers are moving quickly to service architectures that virtualize multiple data centers into what we refer to as a "Data Center without Walls".

This approach is made possible by cloud orchestration software that federates the data centers of both the enterprise customer and cloud service provider such that all compute, storage, and networking assets are treated as a single, virtual pool that optimizes placement, migration, and interconnection of workloads and associated storage. This architecture provides service resiliency and performance regardless of the user's location, while simultaneously creating resource efficiency gains for the cloud service providers. According to analyses conducted with some major operators, a Data Center Without Walls approach can yield resource efficiency gains of 35 percent over isolated provider data center architectures.

SDN enables cloud services to benefit from performance-on-demand

The Data Center Without Walls service architecture is not without network challenges. Application dynamics frequently trigger migration of workloads between enterprise and public clouds, or between clusters of service provider data centers, creating traffic between those two endpoints. Any inadequacies of the network connection in terms of bandwidth, latency, or quality, will have a detrimental effect, either on the practicality and convenience of the operation or on the performance of the application.

Many of the loads and demands placed on the data center interconnect network are transactional - meaning they trigger finite bandwidth consumption in terms of both time and quantity of information, rather than the generation of a long-lived traffic flow. Consider, for instance, self-service customer application operations - application creation, re-sizing, or deletion in the cloud - or certain provider administrative operations, which can cause variability and unpredictability to traffic volumes and patterns. To accommodate this variability in traffic, providers normally would have to over-provision the network to handle the sum of peaks of all traffic - an inefficient and costly approach.

SDN-based Performance-on-Demand

In the future, rather than over-provisioning the network, service providers will employ a Software-Defined Network (SDN) that can be programmed to allocate bandwidth from a shared pool of resources where and when it is needed. The SDN framework consists of the following:

- The infrastructure layer (or Cloud Backbone): The transport and switching network elements forming the inter-data center network

- The network control layer (or SDN Controller): The software that configures the infrastructure layer to accommodate service demands

- The application layer (or Cloud Orchestrator): The service-creation/delivery software that determines required network connectivity

The SDN Controller and Cloud Orchestrator can work hand-in-hand to optimize the service ecosystem as a whole: compute and store resources, plus connectivity across the Cloud Backbone.

In the infographic shown in this article, the SDN Controller is the key enabler for application driven performance-on-demand. Effectively it allows the Cloud Orchestrator to request allocation of networking resources without needing to understand the complexity of the underlying network. The SDN Controller has sight of the entire landscape of all existing connections, anticipated connections, and unallocated resources, making it more likely to find a viable path if one is possible even if nodes or links are congested along the default, ie the shortest route.

Meanwhile, the Cloud Orchestrator at the application layer has full visibility of inter-data center workload requirements and can thus choose destination data centers and schedule transactions to maximize the performance of the cloud service. Through communication with the SDN Controller, it can select the best combination of connection profile, time window and cost.

For example, the Cloud Orchestrator may simply request a connection between specified hosts in two different data centers to handle the transfer of 1 TB with a minimum flow rate of 1 Gb/s and packet delivery ratio of 99.9999% to begin between the hours of 1:00 and 4:00 am. The SDN Controller first verifies the request against its policy database, performs path computation to find the best resources for the request, and initiates the provisioning of those resources. It subsequently notifies the Cloud Orchestrator so that it may in turn initiate the inter-data center transaction.

Analysis shows that delivering such SDN-based performance-on-demand reduces inter-data center capacity requirements by up to 50 percent compared to a network over-provisioned to the sum of all peak demands.

Beyond the substantial CapEx efficiencies, SDN automation simplifies planning and operational practices, reducing the costs associated with these tasks. Programmed policies, not manual "swivel chair" processes, drive service operation selections and sequences.

Coordinating cloud compute and storage activities with Cloud Backbone network allocation smoothes out demand curves, and allows delivery of more services with fewer resources yet with the robustness and performance that enterprises demand.

With approximately 30% of Asia Pacific organizations having adopted some form of cloud computing by 2012 (source: Gartner), the time to investigate the advantages of SDN-based Cloud Backbones and service infrastructures is now.