Special Coverage Series

Network Computing

Special Coverage Series

Commentary

Ethan Banks

SDN Startup Targets Cloud Providers

SDN startup Anuta Networks is launching a controller for IaaS cloud providers to manage physical network infrastructure—switches, routers, load balancers and more—and to automate the provisioning of network services.

The number of startups in the software defined networking space is a testament to the faith industry insiders have in SDN's future.

SDN is a bold notion: Tell the network how to behave using an omniscient, application-driven controller and eschew individual device management. However, in this rapidly evolving way of thinking about the network, just how to achieve network virtualization nirvana is what sets apart vendor approaches.

More Insights

Webcasts

More >>

White Papers

More >>

Reports

More >>

SDN solutions vary greatly, depending on what problem is being tackled. An enterprise's network challenges differ from a WAN service provider's, which differ yet again from an IaaS cloud provider's. That's good for SDN startups; they have room to provide a unique solution without all competing for the same chunk of revenue.

Anuta Networks came out of stealth at Cisco Live in London this January. Anuta's target market is the cloud provider IaaS space.

What is Anuta selling? In a nutshell, Anuta's nCloudX product is a controller able to manage the physical infrastructure many networks already have in place today. In other words, Anuta isn't requiring multitenant cloud providers to install OpenFlow switches, deploy a new type of overlay, or otherwise disrupt the production environment to gain SDN-based orchestration and services management.

Cloud Computing Comparisons: IaaS Providers
16 top vendors. Detailed features matrixes with over 60 data points. Your one-stop shop for an IaaS short list.
See Infrastructure As A Service Comparison »

Its goal is to abstract commonly deployed network routers, switches, firewalls and application delivery controllers, along with well-understood forwarding paradigms like BGP, OSPF, MPLS and VRFs, and hide them behind the nCloudX controller. Cloud operators and end users can then provision the network using templates predefined by the cloud designer that obviate the need for manual provisioning of each network layer.

How It Works

The nCloudX controller runs as a virtual machine on a server in a provider's existing infrastructure. The controller has a northbound REST API that applications use to tell nCloudX about provisioning requirements. Anuta has relationships with several vendors, enabling them to write to its API, including Microsoft's System Center Virtual Machine Manager. The software also plugs in to Quantum, the network component of OpenStack, for network abstraction services.

In the southbound direction, the nCloudX controller programs network devices through a combination of APIs, SNMP and automated CLI interaction. The company says it can integrate with a variety of devices from Cisco, including the Catalyst 6000 and 3000, the Nexus 5000 and 7000, and the ASR 1000. The controller also works with Citrix's NetScaler ADC and load balancers from F5 and Riverbed.

For environments interested in southbound OpenFlow and perhaps overlays for tenant separation, support for both are on Anuta's near-term roadmap.

[ Join us at Interop Las Vegas for access to 125+ IT sessions and 300+ exhibiting companies. Register today! ]

Once nCloudX is in place, there are several engines that an Anuta customer interacts with to design, provision, and monitor their network environment.

• The Service Design Engine builds a series of service-level templates or "network services" that are published.

• The Service Orchestration Engine maps the network service templates to the physical infrastructure.

• The Service Management Engine monitors the physical network infrastructure, and escalates alerts up through network service and tenant containers. Operators and tenants can tell at a glance which of their services are impacted because of a failure in the underlying network.

• The Capacity Management Engine tracks the ability of the underlying network infrastructure to support additional network services through the tracking of CPU, memory, VLANs, VRFs, bandwidth, firewall contexts and so on.

Interestingly, a topology discovery does not have to be completed before designing network services. The process of mapping network services to physical infrastructure is not dependent on the infrastructure actually being there; the mapping can happen later.

The demonstration I received of the nCloudX interface revealed an appealing design aesthetic. One the one hand, working with objects like load-balancers or firewalls was as simple as dragging and dropping icons. For some, that level of abstraction is both sufficient and important to keep their jobs simple.

On the other hand, certain screens take the viewer far more deeply into the nuts of bolts of the network, a necessary element for designers and engineers. As a network engineer, I especially liked the topology screen showing how a particular tenant's services flow through the network infrastructure, right down physical network interfaces.

The Anuta team has taken on a number of complex, error-prone networking tasks and wrapped a tidy controller around them. While Anuta's initial market push is into IaaS, it's easy to imagine applications for nCloudX in academia and large enterprises with multitenant needs, or in the broader enterprise arena where provisioning automation, global policy deployment and application performance monitoring remain pain points largely lacking holistic solutions. Before that happens, however, the company has to demonstrate that it can meet its promises in real-world deployments.



Related Reading



Network Computing encourages readers to engage in spirited, healthy debate, including taking us to task. However, Network Computing moderates all comments posted to our site, and reserves the right to modify or remove any content that it determines to be derogatory, offensive, inflammatory, vulgar, irrelevant/off-topic, racist or obvious marketing/SPAM. Network Computing further reserves the right to disable the profile of any commenter participating in said activities.

 
Disqus Tips To upload an avatar photo, first complete your Disqus profile. | Please read our commenting policy.
 

Editor's Choice

RESEARCH: 2013 Backup Technologies Survey

RESEARCH: 2013 Backup Technologies Survey

Think backups are boring? Not so, say more than 500 IT pros. Most, 60%, use two, three or even more different backup applications, and the percentage encrypting all media has jumped 15 points since 2011.
Get full survey results now! »

Digital Issue: The Standardization Debate

Digital Issue: The Standardization Debate

An IT infrastructure constructed from uniform blocks of hardware and software is easier to manage and secure, and new services can be rolled out fast. But giving business units carte blanche can deliver more flexibility, drive innovation and better meets employee needs. Two IT executives square off in this debate, and almost 400 survey respondents weigh in too.
Get the Digital Issue »

WEBCAST: Avoiding Downtime: How Virtualization Can Help In Times of Trouble

WEBCAST: Avoiding Downtime: How Virtualization Can Help In Times of Trouble

Server and storage virtualization can help keep systems alive even in the face of demand spikes, disasters and other troubles. Attend this webcast to learn how virtualization can maximize application availability, create business continuity options for critical apps, and improve disaster recovery.
Register Today »

Related Content

From Our Sponsor

Implementing Energy Efficient Data Centers

Implementing Energy Efficient Data Centers

Electrical power costs over the life of a data center may exceed the initial cost of the IT equipment. As described in this paper, recognizing the appropriate IT design architecture necessary and being able to quantify the potential electrical savings can significantly increase cost savings over time.

Creating Order from Chaos in Data Centers and Server Rooms

Creating Order from Chaos in Data Centers and Server Rooms

IT Professionals who are challenged with managing a chaotic data center - messy racks, sub-standard floor air distribution and cable sprawl - can now leverage innovative methods for dealing with and eliminating the root causes of disorder. This paper outlines the solutions available to help create an organized data center.

High-Efficiency AC Power Distribution for Green Data Centers

High-Efficiency AC Power Distribution for Green Data Centers

In order to create optimal electrical efficiency and simplified data centers, the use of 240 volt power distribution is highly recommended. This paper describes the various configurations for this distribution architecture as well as the quantified benefits. Note: Applicable to North America only.

Energy Efficient Cooling for Data Centers: A Close-Coupled Row Solution

Energy Efficient Cooling for Data Centers: A Close-Coupled Row Solution

The trend of increased heat densities in data centers has held consistent with advances in computing technology. As power density increased, the degree of difficulty in cooling these higher power loads was also increasing. This article discusses the efficiency benefits of row-based cooling compared to two other common cooling architectures.

Data Center Projects: Standardized Process

Data Center Projects: Standardized Process

As the design and deployment of data centers evolve into more complicated projects, the benefits of a standardized and predictable process are compelling. This paper presents an overview of a standardized, step-by-step process methodology that can be adapted and configured to suit individual requirements, thus reducing costs and eliminating waste.