User Tools

Site Tools


requirements_projects:high_availability_for_vnfs

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
requirements_projects:high_availability_for_vnfs [2014/11/20 12:42]
Qiao Fu
requirements_projects:high_availability_for_vnfs [2014/12/17 06:35] (current)
Murali Birru [Committers and Contributors:]
Line 1: Line 1:
-====High availability for VNFs====+====High availability for OPNFV====
  
-  * Proposed name for the project: ''​High availability for VNFs''​ +  * Proposed name for the project: ''​High availability for OPNFV''​ 
-  * Proposed name for the repository: ''​HA''​ +  * Proposed name for the repository: ''​availability''​ 
-  * Project Categories: ''​(Documentation),​(Requirements),​(Integration & Testing),​(Collaborative Development)''​+  * Project Categories: ''​(Requirements) ''​
  
 ====Project description:​ ==== ====Project description:​ ====
  
-This project is based on the requirement proposal in “HA requirement-Chinamobile.pptx”,​ which is proposed in the initial TSC meeting. The proposal is focus on the high availability requirements ​and some possible development ​of code for the VNFs runing on the OPNFV platform.  In the current networks, normally HA is required for the physical harwaresAnd HA can be fullfilled by 1:1 backup. When it comes to the NFV case, since VNFs are all virtualized on different VMs, HA problem can become complecated. In this project, we address HA requirements and solutions in 3 diffenrent perspectives,​ the hardware HA, the virtual mechine HA and the service HA, to be specific+This project is based on the requirement proposal in “HA requirement-Chinamobile.pptx”,​ which is proposed in the initial TSC meeting. The proposal is focusing ​on the high availability requirements of the OPNFV platform ​regarding ​the Carrier Grade NFV scenariosSome possible development of codes to meet the requirement may also be developed
  
-The most important and difficult is the HA for the services. So for, the possible work this project ​will deal with is as follows: ​+In the current networks, normally ​HA is required ​for the physical hardwareand can be fulfilled by 1:1 backup. When it comes to the NFV case, since VNFs are all virtualized on different VMs, HA problem can become complicated. In this project, we address HA requirements and solutions in 3 different perspectives,​ the hardware HA, the virtual infrastructure HA and the service HA, to be specific. ​
  
-1) What is the requirment ​of HA for carrier grade VNFs?+The hardware HA can be solved by several legacy HA schemes. However, when considering ​the NFV scenarios, hardware failure will cause collateral damage to not only virtual infrastructures but also services running on it. Therefore, fault detection and report ​of HW failure from the hardware to VIM, VNFM and Orchestrator is necessary ​for the HA approaches in OPNFV. ​
  
-2) The requirment of HA for different services might be different. HA criterian ​for different ​services should ​be definedCriterian ​may not just about the reliability,​ but should also take latency, packet loss rate, and etc. into consideration.+Virtual infrastructure ​HA addresses the high availability mechanisms ​for NFV infrastructures and VIMThe HA schemes Openstack now have, which provides ​HA for different ​components for Openstack (such as Nova, Quantum, and etc.), can be categorized into this classHowever, considering the Carrier Grade NFV scenarios, current HA schemes ​may not be enough. Gap analysis for Openstack HA schemes can be done in this project. Based on the analysisfurther improvement ​and development can be done to improve the HA schemes for the virtual infrastructure.
  
-3) How to detect ​service ​failure?+There are several different mechanisms for the service ​HA. In this project, a white paper concluding these different mechanisms will be accomplished. Based on these mechanisms, we can further study what capability and interfaces OPNFV platform shall provide to the services. Therefore, API provided by the OPNFV platform to the services HA schemes should be well defined.
  
-4) How to redirect ​service ​flow to the new service VNF?+In this project, we will first work on the HA requirements of Carrier Grade NFV. The requirements will include the hardware HA, the virtual infrastructure HA, and the service ​HA. HA API will then be well defined according ​to the requirements. Such API may include not only state update but also capability provision from the hardware and platform to the Orchestrator,​ the VNFM and different VNFs, 
  
-5) How ​to accomplish service backup? How to prevent share of HW of service ​and its backup?+For hardware HA, legacy HA mechanisms will be studied. Considering the NFV scenario, API will be defined for the hardware ​to inform the up-layers about its states ​and its capability. Related interfaces may include vl-hw, nf-vi, vi-vnfm, and or-vi. ​
  
-6) Instread of using 1:1 backupis it possible to provide more resource efficient 1:N backup?+For the virtual infrastructure HAgap analyses of HA mechanism will be proposed for the current Openstack. Advance mechanisms of OPNFV HA can be studied and developed. And corresponding Openstack extension can then be developed in the following collaborative development project. ​
  
-7) How to locate ​the failed ​service ​when failure happens at HW or VM?+For the service ​HA, studies about the current service HA schemes can be pulled out. In the meantime, API can be defined for state update and capability provision for the VNFs from the lower layers. Related interfaces may include vn-nf, vi-vnfm and or-vi. 
  
-8) Realization ​of HA for different criterians.+Based on the output ​of this requirement project, corresponding testing may also be worked out about the HA APIs and possible development in the following integration and testing project
  
-9) Interfaces of orchestrator,​ VIM, VNFM when accomplish service HA.+====Scope:​====
  
-10) Interfaces of orchestrator,​ VIM, VNFM when different criterian of HA is choosen ​for each service.+• Propose high availability requirement ​for Carrier Grade NFV.
  
 +• Define HA API requirements. Related interfaces may include vl-ha, vn-nf, nf-vi, vi-vnfm, or-vi.
  
-We proposed the HA for VNFs can be achieved by the HA API offered by the OPNFV platform. In this project, we will first work on the HA requirements and KPI of VNFs. HA API and corresponding ​Openstack ​extention can then be developed to meet the requirements. We will also work on testcases to validate the interoperability of different VNFs with the HA API.+• Gap analyses ​of Openstack HA schemes
  
 +• Whitepaper about service HA scheme study
  
-====Scope:====+====Possible work for the following projects:====
  
-Propose high availability requirement ​and KPI for the NFV platform.+Openstack extension ​and development on HA
  
-HA requirements shall include hardware HA, virtual machine HA, and service HA. +Testcases of OPNFV HA API and development.
- +
-• Contribute ​HA API and corresponding Openstack extentions. +
- +
-• Testcases of HA for the platform with different VNFs based on the requirement.+
  
 ====Dependencies:​==== ====Dependencies:​====
-• Identify similar projects is underway or being proposed in OPNFV or upstream project 
- 
-Related with Openstack 
- 
 • Identify any open source upstream projects and release timeline. ​ • Identify any open source upstream projects and release timeline. ​
 +
 +  Openstack
 +  OpenDaylight
  
 • Identify any specific development be staged with respect to the upstream project and releases. • Identify any specific development be staged with respect to the upstream project and releases.
  
 • Are there any external fora or standard development organization dependencies. If possible, list and  • Are there any external fora or standard development organization dependencies. If possible, list and 
 +
 + None 
  
 ====Committers and Contributors:​==== ====Committers and Contributors:​====
-Name of and affiliation ​of the maintainer:Qiao Fu (Chinamobile)+Names and affiliations ​of the committers: 
 +  
 +Qiao Fu (fuqiao@chinamobile.com)
  
-• Names and affiliations of the committers+Palani Chinnakannan (pals) [pals@cisco.com] 
 +  
 +Huailin Chen [huailin@arraynetworks.net]
  
 +Jun Zhang [zhang.jun3g@zte.com.cn]
 +
 +Stefan Arntzen [Stefan.Arntzen@huawei.com]
 +
 +Yuyijun (Eugene) [yuyijun@huawei.com]
 +
 +Yifei Xue [xueyifei@huawei.com]
 +
 +Hellmann, Gil [gil.hellmann@windriver.com]
 +
 +Jonas Bjurel [jonas.bjurel@ericsson.com]
 +
 +Maria Toeroe [maria.toeroe@ericsson.com]
 +
 +Tamas Zsiros [tamas.zsiros@ericsson.com]
 +
 +Murali Birru [murali.yadav@calsoftlabs.com]
 +
 • Any other contributors • Any other contributors
  
  
 ====Planned deliverables==== ====Planned deliverables====
-o HA requirement documents ​+HA requirement documents ​
  
-o HA API specification ​+HA API specification ​
  
-o Testcases for HA in OPNFV platform+• Openstack ​HA gap analyses
  
-o Test reports for HA+• Whitepaper of service ​HA schemes  ​
  
 ====Proposed Release Schedule:​==== ====Proposed Release Schedule:​====
Line 77: Line 98:
  
 • Will this align with the current release cadence • Will this align with the current release cadence
 + May not be as part of the 1st release ​
requirements_projects/high_availability_for_vnfs.1416487351.txt.gz · Last modified: 2014/11/20 12:42 by Qiao Fu