Welcome!

SDN Journal Authors: Yeshim Deniz, Elizabeth White, Liz McMillan, Pat Romanski, TJ Randall

Related Topics: SDN Journal, Microservices Expo, Containers Expo Blog, Agile Computing, @CloudExpo, @DXWorldExpo

SDN Journal: Article

An Intelligent Network Is Critical to Achieve Optimal Cloud Performance

Making the promise of cloud services a reality

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.

Gartner predicts that the global public cloud computing market is set to reach US$131 billion in 2013, up from $111 billion last year. To serve this market with the performance, security and reliability needed, cloud providers are moving quickly to build a virtualized multi-data center service architecture, or 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 so that all compute, storage, and networking assets are treated as a single, virtual pool with optimal placement, migration, and interconnection of workloads and associated storage. This "data center without walls" architecture gives IT tremendous operational flexibility and agility to better respond and support business initiatives by transparently using both in-house and cloud-based resources. In fact, internal studies show that IT can experience resource efficiency gains of 35 percent over isolated provider data center architectures.

However, this architecture is not without its challenges. The migration of workload between enterprise and public cloud creates traffic between the two, as well as between clusters of provider data centers. In addition, transactional loads and demands placed on the backbone network, including self-service customer application operations (application creation, re-sizing, or deletion in the cloud) and specific provider administrative operations can cause variability and unpredictability to traffic volumes and patterns. To accommodate this variability in traffic, providers normally would have to over-provision the backbone to handle the sum of these peaks - an inefficient and costly approach.

Achieving Performance-on-Demand
In the future, rather than over-provisioning, service providers will employ intelligent networks that can be programmed to allocate bandwidth from a shared pool of resources where and when it is needed. This software-defined network (SDN) framework consists of the infrastructure layer - the transport and switching network elements; the network control layer (or SDN controller) - the software that configures the infrastructure layer to accommodate service demands; and the application layer - the service-creation/delivery software that determines required network connectivity - e.g., the cloud orchestrator.

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

The logically centralized control layer software is the lynchpin to providing orchestrated performance-on-demand. This configuration allows the orchestrator to request allocation of those resources without needing to understand the complexity of the underlying network.

For example, the 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 a.m. and 4:00 a.m. The SDN controller first verifies the request against its policy database, performs path computation to find the best resources for the request, and orchestrates the provisioning of those resources. It subsequently notifies the cloud orchestrator so that the orchestrator may initiate the inter-data center transaction.

The benefits to this approach include cost savings and operational efficiencies. Delivering performance-on-demand in this way can reduce cloud backbone capacity requirements by up to 50 percent compared to over-provisioning, while automation simplifies planning and operational practices, and reduces the costs associated with these tasks.

The network control and cloud application layers also can work hand-in-hand to optimize the service ecosystem as a whole. The network control layer 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 shortest route.

The cloud orchestrator at the application layer also has full visibility of inter-data center workload requirements and can choose destination data centers and schedule transactions to maximize the performance of the cloud service. Through communication with the network control layer, it can select the best combination of connection profile, time window and cost.

Intelligent Networking + Cloud for a Data Center Without Walls
Enterprise dependence on public cloud services is poised for the fast track; however enterprise networks between data centers, and from data centers to the cloud, need to evolve to make the promise of cloud services a reality. Using a "data center without walls" architecture which employs SDN-methodology to link physical, virtual and cloud resources, can help enterprises smooth out demand curves and deploy new services and applications with the robustness and performance that business requires.

More Stories By Jim Morin

Jim Morin is a Product Line Director working in Ciena’s Industry Marketing segment. He is responsible for developing and communicating solutions and the business value for Ciena’s enterprise data center networking and cloud networking opportunities. Prior to joining Ciena in 2008 he held roles in business development and product management for several high technology storage and networking companies in Minneapolis.

Jim holds an MBA from the University of St. Thomas and a BA from the University of Notre Dame. He recently served on the Commission on the Leadership Opportunity in US Deployment of the Cloud (CLOUD2).

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


CloudEXPO Stories
With more than 30 Kubernetes solutions in the marketplace, it's tempting to think Kubernetes and the vendor ecosystem has solved the problem of operationalizing containers at scale or of automatically managing the elasticity of the underlying infrastructure that these solutions need to be truly scalable. Far from it. There are at least six major pain points that companies experience when they try to deploy and run Kubernetes in their complex environments. In this presentation, the speaker will detail these pain points and explain how cloud can address them.
The deluge of IoT sensor data collected from connected devices and the powerful AI required to make that data actionable are giving rise to a hybrid ecosystem in which cloud, on-prem and edge processes become interweaved. Attendees will learn how emerging composable infrastructure solutions deliver the adaptive architecture needed to manage this new data reality. Machine learning algorithms can better anticipate data storms and automate resources to support surges, including fully scalable GPU-centric compute for the most data-intensive applications. Hyperconverged systems already in place can be revitalized with vendor-agnostic, PCIe-deployed, disaggregated approach to composable, maximizing the value of previous investments.
When building large, cloud-based applications that operate at a high scale, it's important to maintain a high availability and resilience to failures. In order to do that, you must be tolerant of failures, even in light of failures in other areas of your application. "Fly two mistakes high" is an old adage in the radio control airplane hobby. It means, fly high enough so that if you make a mistake, you can continue flying with room to still make mistakes. In his session at 18th Cloud Expo, Lee Atchison, Principal Cloud Architect and Advocate at New Relic, discussed how this same philosophy can be applied to highly scaled applications, and can dramatically increase your resilience to failure.
Machine learning has taken residence at our cities' cores and now we can finally have "smart cities." Cities are a collection of buildings made to provide the structure and safety necessary for people to function, create and survive. Buildings are a pool of ever-changing performance data from large automated systems such as heating and cooling to the people that live and work within them. Through machine learning, buildings can optimize performance, reduce costs, and improve occupant comfort by sharing information within the building and with outside city infrastructure via real time shared cloud capabilities.
As Cybric's Chief Technology Officer, Mike D. Kail is responsible for the strategic vision and technical direction of the platform. Prior to founding Cybric, Mike was Yahoo's CIO and SVP of Infrastructure, where he led the IT and Data Center functions for the company. He has more than 24 years of IT Operations experience with a focus on highly-scalable architectures.