Abstract
Network function virtualization (NFV) is a concept aimed at achieving telecom grade cloud ecosystem for new-generation networks focusing on capital and operational expenditure (CAPEX and OPEX) savings. This study introduces empirical throughput prediction model for the virtual network function (VNF) and network function virtualization infrastructure (NFVI) architectures based on Linux kernel. The model arises from the methodology for performance evaluation and modeling based on execution area (EA) distribution by CPU core pinning. EA is defined as a software execution unit that can run isolated on a compute resource (CPU core). EAs are derived from the elements and packet processing principles in NFVIs and VNFs based on Linux kernel. Performing measurements and observing linearity of the measured results open the possibility to apply model calibration technique to achieve general VNF and NFVI architecture model with performance prediction and environment setup optimization. The modeling parameters are derived from the cumulative packet processing cost obtained by measurements for collocated EAs on the CPU core hosting the bottleneck EA. The VNF and NFVI architecture model with performance prediction is successfully validated against the measurement results obtained in emulated environment and used to predict optimal system configurations and maximal throughput results for different CPUs.
Highlights
Modeling of network function service performance was a straightforward task determined by welldefined test scope during verification activities in telecom vendors’ laboratories for a long time
We presented the results for different network interface card (NIC) settings of interrupt coalescing (IC), and we made the relation of the observed throughput results to new API (NAPI) processing key performance indicators (KPIs)
This study proposed a VNF and network function virtualization infrastructure (NFVI) architecture model with performance prediction and bottleneck determination for optimal virtual network function performance analysis
Summary
Modeling of network function (physical) service performance was a straightforward task determined by welldefined test scope during verification activities in telecom vendors’ laboratories for a long time. This study proposes VNF and NFVI architecture model with performance prediction based on measurements on the commodity multicore system and by applying model calibration technique. This is the first study proposing VNF throughput prediction model validated in emulated environment. This study extends the work in [2] to consider processing in VMs. The proposed methodology, in accordance with empirical EA bottleneck determination, is used to build a set of optimal modeling system configurations for the target VNF and NFVI architectures per given number of CPU core resources. The defined VNF and NFVI architecture model with performance prediction is further used to predict the throughput performance and provide optimal VNF/NFVI system dimensioning for any given number of CPU cores
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
