Cloud data centers today are beginning to use COTS hardware that comprise platforms with a variety of technologies that offer some basic computing functions as well as performance and higher efficiency functions.With the advent of NFV, network services that used to be deployed on proprietary costly hardware are now deployable quickly and flexibly on COTS hardware platforms. The higher performance obtained using dedicated proprietary hardware can also be achieved in a NFV platform (comprised from COTS) by configuring and/or fine tuning the platform capabilities such as SR-IOV, CPU Pinning, Huge Pages and NUMA topology. In this session, we will discuss how the above mentioned performance parameters could be specified as a requirement per Virtualized Network Function (VNF) and have the Tacker NFV Orchestrator ensure that the VNF is placed on the optimal NFV platform that meets the specified performance requirements.
The Telecom/NFV track is brought to you by OpenStack and OPNFV.
The attendees will walk away with:
- An overview/understanding of how SR-IOV, CPU Pinning, NUMA Topology and Huge Pages can help a VNF placed on NFV platforms obtain performance that is comparable to a dedicated hardware appliance
- How to specify performance requirement configurations for a VNF in Tacker VNFD Templates
Registration Opening Soon
Registration will open for you to purchase passes for the OpenStack Summit very soon!
We're excited to see you there!
