- Validate and benchmark VMware vCenter and OpenStack-based private clouds
- Testbed for single compute node or large environments with many racks
- Individually validate the compute / network / storage performance dimensions
- Measure VIM performance with custom VM instantiation / VM termination test methodology
- Leverage subscription licensing for low startup cost and flexibility of pay-as-you-grow model
- Predefined test scenarios for NFVI validation
- Validate from a VNF or network perspective through real workload and powerful traffic emulation
- Includes industry-proven workload emulation from the OPNFV Yardstick portfolio
- Measure scheduler capability to isolate the good workloads and the bad noisy neighbors
- Measure the performance impact of VNF instantiation and VNF workloads in the presence of (noisy) neighbors
Problem: Ensuring Virtualized Networks Are Carrier-Grade
Network functions virtualization (NFV) is a key building block for 5G mobile networks, with network slicing or mobile edge computing (MEC) becoming possible. In the NFV architecture, various virtual network functions (VNF) run on a common NFV Infrastructure (NFVI) and are controlled by a common management and orchestration (MANO) layer.
Carriers must ensure the new network is capable of providing quality services, but the performance of the entire system is only as good as the weakest link. Of primary concern is dimensioning and configuring the NFVI properly to sustain the workloads required by the VNFs. Validation is key to successful NFVI implementation, but it becomes challenging in such a complex multi-resource system and as it interacts with the rest of the NFV architecture elements.
Solution: NFVI Validation Through Real VNF Workload Generation
Cloud Peak is a web application used to validate the functionality and benchmark the performance of virtualized network infrastructures. By deploying real workloads on top of the NFVI, Cloud Peak provides key insights into the capability to sustain the required VNF workloads.
Using an innovative methodology that decomposes system resources into the elementary compute, network, and storage dimensions, the application isolates configuration issues and performance bottlenecks. Test results are compared to preconfigured performance levels, which provides an easy to understand Pass / Fail result, displayed through an intuitive real-time user interface. From testing a single small server, to validating a whole infrastructure with hundreds of nodes, Cloud Peak can scale to match the capacity of the system under test.