Be sure to check out the full, accompanying article, "Virtualizing Everything On The Network."
To gain better service flexibility and scalability, telecom service providers are preparing to move away from traditional architectures and toward network function virtualization (NFV). However, before beginning the transition, operators must ensure their NFV platforms provide the same, dependable level of carrier-class reliability expected by customers.
Carrier-grade reliability, an essential feature of telecom networks, represents a service philosophy as well as a metric: It guarantees network uptime and availability 99.999% (six-nines) of the time, which translates to no more than 32 seconds of downtime per year. Compare that to enterprise-class IT applications, which only provide three-nines of reliability, or nearly nine hours of downtime per year.
Fortunately, NFV at carrier-grade levels is possible. However, it’s hard to attain because so many software elements across the entire NFV architecture must achieve six-nines. To succeed, an NFV platform should be designed from the ground up to meet the following requirements for the telecom audience:
Availability: The network must provide virtual-machine (VM) redundancy over a geographic range of at least 500 km to allow for continued operation in the event of a natural disaster. When faults occur, the VM infrastructure must recover in less than 500 ms. The network should not drop calls or lose data during failovers.
Security: Telecom networks by nature have stringent security requirements. All observable traffic in a 4G network must be encrypted and visible user data can’t be stored in the system. For an NFV data center or cloud deployment, operators will need to implement multi-tenant isolation and security to ensure that subscribers can’t access one another’s traffic or data. Also, the network must fully implement AAA security protocols (authentication, authorization, and accounting) to prevent unauthorized access, hacking, or terrorist attacks.
Performance: The network must achieve both high throughput and very low latency for critical real-time applications. For virtualization to be feasible, the most demanding customer-premise-equipment (CPE) and access functions require deterministic interrupt latency of 10 µs or less. Live migration of VMs must occur with an outage time under 150 ms.
Management: A carrier-grade system must both prevent and eliminate unscheduled as well as planned downtime windows for network maintenance. It also has to support hitless software upgrades in addition to patches, while backup and recovery must be fully integrated with the platform. Furthermore, it’s mandatory that full support for these capabilities be implemented to interface to the existing operations support systems (OSS) and business support systems (BSS) software.
With careful planning and collaboration across the right mix of ecosystem players, such as the NFV-focused Titanium Cloud ecosystem, service providers can build these capabilities into their NFV deployments. They can then move ahead with the confidence to meet intended objectives for NFV while maintaining a satisfied customer base. Without such planning, they risk increased churn and losing high-value customers, which could offset the many business benefits derived from NFV.