These are exciting times in the industry. I personally am enjoying how two parts of my life, networking and computing, are coming together in interesting ways. The onset of NFV is driving this change, bringing the previously separate disciplines together: telecom network services and general-purpose compute resources.
This will mean learning some new tricks on the telecom side to manage network performance and service assurance. For years, each has had its own separate methods and solutions for monitoring performance. But with networks now running on general-purpose compute resources, we need to unify the systems to effectively monitor and manage network services delivery. Several approaches can be taken.
For starters, as we deploy the new NFV infrastructure, new systems will be deployed to run them (e.g., MANO). One approach is to incorporate new monitoring capabilities within and around these new systems.
We can also utilize the systems traditionally used for managing general-purpose compute resources such as enterprise data center and application monitoring systems.
Finally, since the purpose of NFV is to move traditional network functions from purpose-built network equipment (e.g. routers, switches, firewalls) onto the general-purpose compute resources, we can utilize the systems that have traditionally been used to manage network functions.
Because telecom networks generally maintain higher levels of reliability (five 9s) our concern is that issues arising in one area, for example, on general-purpose compute resources, including hardware, virtual machines and operating systems, do not have a direct negative affect on the overall performance of a network service. We have to ensure that these resources meet the same high standards as traditional services, which are run on purpose-built network devices optimized and stabilized to deliver extremely high levels of availability, failover and rerouting.
At Nokia, we employ our NPM VitalSuite software to provide the first line of support for this new challenge. We leverage its extensive capability to monitor and report on virtual compute resources from the physical layer up to the application layer. This enables us to take local action within the NFV domain and federate KPIs northbound to our Service Quality Manager (SQM) to provide end-to-end service assurance across hybrid services.
Fortunately, this transformation to NFV is happening in a phased approach. Portions of the network environment will change to NFV, while other portions continue to run on purpose-built network devices. This approach means that we will all be working in hybrid network environments for an extended period of time. That gives us plenty of time to learn from each other.
The moral of the story is that as we transform parts of the network to NFV, we need to incorporate these general-purpose compute resource infrastructures and the tools they’ve traditionally used to manage performance. By bringing them together with traditional purpose-built network devices and systems, we should be able to continue to monitor and manage these hybrid environments to the appropriate perfomance levels. Tools, such as the Nokia NPM VitalSuite, will be part of the overall solution — just some of the new tricks we will need to learn as we bring these different domains together.
Learn more about how the Nokia Service Assurance suite can help you to operationalize your digital transformation.
Share your thoughts on this topic by replying below – or join the Twitter discussion with @nokianetworks using #NFV #ServiceAssurance #virtualization