As seen in InterComms Issue 29 2017. See the full article in InterComms here.
Q: With changes from traditional static networks to today's mixed and multi-layer, dynamic infrastructures, what problems are network managers facing when introducing new technologies like SDN, NFV and software-defined data centers?
A: Networks are becoming much more complex and dynamic. Traditional networks have historically been statically deployed, accessed from a fixed location and essentially left alone for the duration of their lifetime. Today, a number of factors, on top of the introduction of new technologies and increasing software-definition, are unquestionably impacting how network managers operate and maintain their networks. Service workloads are being deployed on-demand at a much quicker rate - you could say at the speed of software, and their end users are now accessing these services from a greater variety of locations, including home offices, corporate premises and even public locations, creating a much more complex network environment to handle.
But regardless of this increasing complexity, end users and employees are expecting services to work flawlessly and as soon as they request them - their expectations and demands are increasing, just like the increasing network complexity.
In order to successfully manage these dynamic NFV environments and achieve this goal of agile and assured service creation and delivery, network managers need to ensure that they are effectively testing and monitoring these services that they are rapidly provisioning and delivering. How do you do this? With software-based active testing and monitoring that is automated and integrated as an elemental part of the service provisioning and fulfillment processes.
Q: How has automation of configuration and testing helped with allowing successful deployment and running of new network services?
A: Using an orchestrator to configure and change new and existing services on the fly has certainly impacted the rate at which network managers can deliver new services to their users, as well as their ability to automate many steps and processes that were previously very time-consuming and error-prone. A key success factor for delivering new services quickly and delivering them right the first time, though, is the addition of the programmable active testing and monitoring component.
Thorough service activation tests are important as there are many moving parts in this new complex network landscape, and many times there are components of the end-to-end service chain that automation does not touch, such as off-net connections or unmanaged network domains. In addition, dynamic behavior such as varying network load can also influence end-to-end service quality. If you are increasing the rate of new service deployment through automation but you are delivering services that are not working once they are delivered, are you positively improving your network operations practices and your organization’s operational efficiency? I wouldn’t say so.
Automating active testing and monitoring does lead to a number of very tangible and immediate benefits, including decreased trouble tickets as you are testing provisioned services prior to customers using them and realizing that they were not provisioned correctly. Our customers also see a decrease in their ticket handling times, dispatches of field technicians, and in service providers’ cases, churn. Their service quality insights and time to deployment of new, assured services are greatly improved, even without the use of an orchestrator, and ultimately, user satisfaction is increased as well.
Q: Could you talk through Netrounds’ Control Center and Test Agents - how they operate and the unique benefits they bring to network managers?
A: The Netrounds system is made up of two main components - the Netrounds Control Center and our Netrounds Test Agents with active traffic generation capabilities. The Netrounds Control Center provides a graphical designer environment for network managers to build their service activation testing templates and other automatable testing and monitoring sequences. These templates can then be triggered using the NETCONF and YANG API to achieve closed-loop automation. The Netrounds Control Center GUI environment also presents test reports and real-time key performance indicators (KPIs) of actual end-user experience through dashboards and drill-down charts.
The Netrounds Test Agents are software based and run as virtual machines on common hypervisors or directly on x86 bare-metal. For full automation, an orchestrator can dynamically deploy Test Agents in virtual environments at the time of service configuration, as we discussed before. Think of these Test Agents as your virtual end users located on the network. They can generate authentic traffic and interact with real services, as well as request and download content, to analyze detailed, real-time measurements across multiple applications, services, and interfaces - from Layer 2 to Layer 7. They support concurrent measurement of network performance (UDP, TCP, Y.1731, and TWAMP), IPTV and OTT video, and the Internet (HTTP, ping, and Speedtest), as well as VoIP and Session Initiation Protocol (SIP) telephony. All Test Agents are centrally managed and updated from the Netrounds Control Center using an automated call-home mechanism.
Q: Netrounds has some major clients, like Telenor and Telia. How have they benefited from using the products in real world situations?
A: Our customers today are represented by about fifty percent enterprises and fifty percent large network operators and service providers. Telenor, one of our larger operator customers, has been extremely progressive in their use of the Netrounds platform and how it interacts with their orchestration system and some of the surrounding OSS components. We are also currently working with them in the Telenor Research group to push the envelope a bit further in terms of closed loop automation and triggering Netrounds to perform tests in order to optimize end-to-end service quality as experienced by their end users.
Telia has really benefited from Netrounds’ ease of use in particular. As we offer Netrounds as a SaaS solution -- in addition to private cloud deployments -- Telia found it very easy to get started quickly with only a minimum of training.
With our cloud-based SaaS offering, our customers can focus on their core business and revenue-generating services rather than worrying if those services are working correctly and delivering the value they promised. I can actually quote one of our users at Telia as saying that “we chose Netrounds because it’s miles ahead of the competition in terms of value for the money.” Another big benefit.
Q: You are strategically aligned with Cisco Systems, and in particular, integrating your platform with Cisco’s CSP 2100 could you discuss this?
A: This new joint solution is a turn-key and scalable NFV solution for enterprises aimed at lowering cost, time and technical expertise barriers that are hindering them from deploying virtual solutions. It allows enterprises to easily begin taking advantage of the automation and operational efficiency benefits that NFV has promised. It involves Netrounds deployed on Cisco’s Cloud Services Platform 2100 with Cisco’s Network Services Orchestrator (NSO).
The Cisco CSP 2100 is built on the Linux Operating System and tightly coupled with Cisco UCS x86 servers with the latest drivers and firmware so you have a convenient appliance form factor that provides the agility of software along with the benefits of performance-optimized hardware, and then you have added in the market-leading orchestrator, Cisco NSO, together with our Netrounds active testing and monitoring system for automated verification of network service and quality. It really is an all-in-one solution that can be racked, switched on, and immediately start impacting your business.
For further information or to request a demo, please email email@example.com.
About the author
Stefan brings over 25 years of experience in automation and the OSS landscape to Netrounds as a Product Strategy Advisor. Today, Stefan is also an Advisor to Data Ductus, a System Integrator specialist in the United States and Sweden. Prior to joining Netrounds, Stefan was the Director of Product Strategy at Tail-f Systems, followed by a Principal Engineer working on NSO Strategy at Cisco after Cisco’s acquisition of Tail-f in 2014. Stefan holds a Ph.D. in Network Management from Luleå University of Technology and a M.Sc. in Computer Science from Linköping University in Sweden.