Communication service providers (CSP) have all kinds of customers. Some of them, especially large enterprise and business customers, require premium connectivity services with guaranteed delivery, and they are willing to pay for it. Typically, premium enterprise services include high capacity, low latency, and full geographic diversity guarantees. A few examples are financial services connectivity, connectivity from primary to backup sites, dual homing to mobile base stations, and connectivity of the enterprise to public cloud.
While premium services generate higher revenue for the CSP, they also come with higher commitment, known as the Service Level Agreement or SLA. Ensuring the service can be established at the right SLA means the CSP must know what resources are available and where. Mistakes in this process require ordering additional gear and result it prolonged lead times before the service can be established – causing customers to pic another CSP (order fallout).
Once the service is established, the SLA must be assured. Having to pay SLA penalties can erode the profit of these lucrative services. To assure SLAs and avoid penalties, service providers rely mostly on overprovisioning. This is expensive but viewed as necessary because the lack of closed-loop control over the network does not allow them to fully guarantee the SLA along the entire service path and the entire service lifecycle. While overprovisioning can resolve bandwidth issues, it cannot improve latency or guarantee better diversity.
This puts service providers in a bind because the tools at hand are not ideal. Inventories are often outdated or erroneous. This lack of accurate network topology data is compounded by the lack of multi-layer tools to monitor and control SLAs all the way down to the physical fiber route. As a result, service issues are difficult to identify and fix before the SLA is violated and the customer is affected. While having to pay SLA penalties incurs an immediate loss, even more serious impact may show up later in the form of loss of customers, and brand erosion.
There’s a better way to guarantee SLA fulfillment.
- It starts with accurate network inventory and service path data. Not just at one layer of the network, but at all layers of the network – simultaneously. Not just for one domain, but for all domains. Not just for one type of vendor equipment, but for all vendor equipment. The reason is simple. You can’t control what you can’t see or don’t know about.
- It continues with Software Defined Networking (SDN) that enables full automation of service provisioning and SLA verification, and allows CSPs to finally “close the loop” by automatically fixing services that drop below contracted service levels.
While SDN is still work-in-progress for many service providers, it pays to know what is possible once SDN is deployed in the network.
In an SDN-enabled network, the Sedona NetFusion Network Intelligence and Automation Platform makes your entire network the inventory of record. A live model shows you every nook and cranny of the network and makes it easy to access and manipulate. That means service providers can configure, provision and control network services across all layers, all network domains, and all vendor equipment. All from a single dashboard.
Armed with accurate and complete network data, service providers can automate optimization of the entire network – shifting services to make room for new deployments, adding capacity as needed, altering service paths to make them more diverse and resilient, and provisioning services from end to end of their lifecycle.
Let’s take a closer look at how it works.
Left: the current best-effort approach to Premium Services Fulfillment is expensive and fraught with shortcomings. Right: the data-driven approach automates the process, making SLA fulfillment more efficient, reliable, and less costly
Once an order is received, Sedona NetFusion Network Intelligence and Automation Platform checks with the live network inventory data to verify that the network resources needed to deliver the SLA are indeed available. If required resources are available, the service can be provisioned immediately. If not, additional cards at endpoints or along the way may need to be ordered and installed first.
Assuming all is in order, the new service is provisioned automatically, which ensures a fast and accurate process. The automated process can perform actions across multiple domains and sometimes even multiple layers. For example, perhaps more capacity between routers is needed to route the service along a preferred path, which may require one or more changes to the underlying optical switch infrastructure.
Automation never stops as the system continuously monitors the service to ensure its SLA is being fulfilled at all times. Proactive alerts notify the operations team when service level issues occur and provide several tools to help troubleshoot and fix the problem. For example, SLA issues may be corrected by rerouting the service to less congested path or by increasing capacity on IP network links.
Sedona Systems is helping service providers advance their SDN deployments and automate premium services monitoring and fulfillment, so they can offer competitive SLAs and ensure that they are met