Telecom IQ
Showing results for 
Search instead for 
Did you mean: 

Inside your NFV Platform Choices – Choosing a Virtual Infrastructure Manager


Inside your NFV Platform Choices – Choosing a Virtual Infrastructure Manager


ShowingHPENFVSystem.jpgYou are a Communications Service Provider operating fixed or mobile infrastructure and you have recognized the advantages of bringing cloud technology into your network.   You’ve decided to begin your transition to virtualized network infrastructure.  

You want to make this transition using open standard interfaces where possible to avoid the kind of vendor lock-in and vendor “silo’s” you experienced with your telecom infrastructure in the past. However, you also want the choice to buy your VIM layer integrated with NFV-I and Orchestration to avoid system integration overheads.


You’ve embraced the ETSI reference architecture for NFV and want to take advantage of COTS hardware and software including Open Source solutions for their standard APIs, broad community support, and enhanced features. But you also required higher SLAs and better performance than standard IT infrastructure provides.

As you begin to look at your NFV platform, including NFV Infrastructure, Virtual Infrastructure Management, and Orchestration (referring to ETSI terminology) as well as your Virtual Network Functions, you may choose to acquire a complete NFV platform, or individual components.   Regardless, a critical element in achieving openness, while providing cloud agility and efficiency, is your choice of Virtualization Infrastructure Manager and compute/network virtualization technologies.

A VIM is a cloud management system, and with its underlying compute and network virtualization software, provides the software platform for all of your network functions. To achieve your NFV goals, you need a VIM that:

  • Provides affordable, carrier levels of SLAs, performance and manageability,
  • Uses Open APIs and standards to avoid lock-in
  • Based on open source where possible
  • Is pre-integrated and pre-tested out of the box with a range of compute and networking options, as well as Orchestration and OSS 

Carrier Grade Availability and performance vs enterprise grade OpenStack

To be carrier grade your VIM and virtualization layer must work together to ensure fast fault detection (sub-second) across the Virtual machines, platform and networking.  In case of issues fast live migration must be supported.    Line rate throughput must be supported to ensure efficiency of the compute infrastructure and adequate performance of I/O and network heavy applications.

Carrier Grade Manageability vs enterprise grade OpenStack

Unlike for enterprise grade OpenStack, CSP users expect full automation, from the NFV VIM deployment itself to the ongoing maintenance and provisioning of network services.   OSS/BSS integration is critical to support flow through provisioning and automated, policy-driven fault handling.

Openness and Bundling

CSPs want both the openness of standard APIs and open source with the confidence that, if desired, a single vendor can handle the entire NFV platform.   Openness can be measure by which APIs and open source solutions supported, but also by the number of application vendors and Network Equipment providers that have tested with or built on the NFV Platform.  Which NFV Platform components can be purchased bundled with/pre-integrated with, the NFV platform?  For the VIM, Orchestration, SDN, Virtualization, Physcial Infrastructure, and of course, VNFs from various vendors should be considered.


The HPE Approach

To meet these needs, Hewlett Packard Enterprise has developed HPE Helion OpenStack Carrier Grade (HCG), based on our Helion OpenStack Enterprise Edition, with our own and partner additions/enhancements.

HCG includes our Carrier Grade OpenStack, with a set of compute and network virtualization options that have been selected or built to consistently provide the reliability, performance and manageability CSPs need.  These include a carrier-grade KVM and Linux and enhanced virtual switching and a carrier SDN solution for networking.  Of course other versions of these can also be supported.  

We've invested heavily in both bundling of our own and partner NFV platform products, and the testing of VNFs.   More than 30 vendors have already tested on our platform in our 5 labs worldwide, with more than that number coming into testing.

We’ve worked with ACG to characterize the benefits of our carrier grade features against generic enterprise OpenStack. We also see results from our current customers versus existing approaches. ACG has calculated a 19% TCO benefit thanks to our carrier grade performance, and up to $1.2B of savings thanks to the increased reliability and $55K/year in improved automation. From current production environments in NFV we know that our OSS integration and orchestration reduces time to market for new services dramatically and time to activate these services for customers can be reduced from 30 days down to 5 minutes.

As more deployments of NFV on our HCG go live we will publish more business metrics to help you in your own choices. Meanwhile – see the ACG paper here, and find out more on HCG here.

0 Kudos
About the Author


Read for dates
HPE Webinars - 2019
Find out about this year's live broadcasts and on-demand webinars.
Read more
Read for dates
HPE at 2019 Technology Events
Learn about the technology events where Hewlett Packard Enterprise will have a presence in 2019.
Read more
View all