HPA requirement are not being defined by IFA. The IFA model simply defines the definition structure – what information elements are used to contain different parts of the HPA requirements. The HPA requirement names are driven by hardware vendor, in our case – ARM and Intel…
From: [email protected] [mailto:[email protected]] Sent: Thursday, March 1, 2018 12:50 PM To: Vul, Alex <[email protected]> Cc: [email protected]; [email protected]; [email protected] Subject: 答复: RE: Re: [onap-discuss]答复: [modeling] Comparison of R2 proposed IM classes and IFA011 Hi Alex IFA does not define the HPA specific parameters until now, but the HPA definition or implimention of ONAP model needs. What I mean is that which specific HPA parameters are need to be modeled in the R2. We need the inputs. In the ONAP, we need to discuss how to define it and the projects need to know how to impliment it. BR Maopeng 原始邮件 发件人：Vul,Alex <[email protected]<mailto:[email protected]>> 收件人：张茂鹏10030173; 抄送人：[email protected]<mailto:[email protected]> <[email protected]<mailto:[email protected]>>[email protected]<mailto:[email protected]> <[email protected]<mailto:[email protected]>>[email protected]<mailto:[email protected]> <[email protected]<mailto:[email protected]>> 日 期 ：2018年03月01日 19:26 主 题 ：RE: Re: [onap-discuss]答复: [modeling] Comparison of R2 proposed IM classes and IFA011 Hi Maopeng, First, observe that “HPA Enablement” is not a project, but a functional requirement for R2 that is being implemented across multiple projects. This has been discussed and approved by the TSC and affected projects have subsequently passed the M1 and M2 milestones. The current plan is to support the set of HPA capabilities available via OpenStack APIs. We plan to use the vCPE VNFs that were provided by Intel for testing purposes. Existing lab hardware can be used for this. We are looking to convert the existing VPP VNF’s, that were built by Intel for vCPE testing from HEAT to TOSCA and use them to test HPA. Existing test flows can be used. Intel will provide the hardware to do the testing. Also note that HPA enablement is meant to enable future 5G work in R3. From this perspective, this is no different than the ONAP work to support PNFs. Finally, HPA functionality is modeled based on the ETSI NFV IFA011 model changes introduced and approved in v2.3.1. These model changes were introduced based on actual VNF developer engagements with Intel, as well actual model usage. My expectation is that our R2 information and data models contain the right elements and attributes to support the HPA implementation in R2. Kind regards, Alex Vul Intel Corporation From: [email protected]<mailto:[email protected]> [mailt:[email protected]] Sent: Thursday, March 1, 2018 10:25 AM To: Vul, Alex <[email protected]<mailto:[email protected]>> Cc: [email protected]<mailto:[email protected]>; [email protected]<mailto:[email protected]>; [email protected]<mailto:[email protected]> Subject: 答复: Re: [onap-discuss]答复: [modeling] Comparison of R2 proposed IM classes and IFA011 Hi Alex As a model contributor, join the discussion. Could you give the specific HPA parameters and completed usecases in R2? Does the LAB need to provide some specific hardwares to test these features? If we model the HPA parameters and implemented in R2, I think some real implemented inputs are needed. Thanks Maopeng 原始邮件 发件人：Vul,Alex <[email protected]<mailto:[email protected]>> 收件人：yangxu (H) <[email protected]<mailto:[email protected]>>jessie jewitt <[email protected]<mailto:[email protected]>>[email protected]<mailto:[email protected]> <[email protected]<mailto:[email protected]>> 日 期 ：2018年03月01日 16:48 主 题 ：Re: [onap-discuss]答复: [modeling] Comparison of R2 proposed IM classes and IFA011 _______________________________________________ onap-discuss mailing list [email protected]<mailto:[email protected]> https://lists.onap.org/mailman/listinfo/onap-discuss Jessie, Xu There are a few things still missing from the agreed upon model. We need to add in items related to the HPA support. HPA is a functional requirement being implemented in R2. We need to ensure that the agreed upon model does not preclude the implementation of HPA. Thank you, Alex Vul Intel Corporation From: [email protected]<mailto:[email protected]> [mailto:[email protected]] On Behalf Of yangxu (H) Sent: Thursday, March 1, 2018 9:29 AM To: jessie jewitt <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> Subject: [onap-discuss] 答复: [modeling] Comparison of R2 proposed IM classes and IFA011 Hi Jessie, For the agreement, please look at https://wiki.onap.org/display/DW/Design+Time+Model+Clean+Version. The differences with IFA011 are shown in orange. As for “vnfProductName”, the agreement is to have the same name as IFA011 for the time being. Best regards, Xu 发件人: [email protected]<mailto:[email protected]> [mailto:[email protected]] 代表 jessie jewitt 发送时间: 2018年2月28日 23:45 收件人: [email protected]<mailto:[email protected]> 主题: [onap-discuss] [modeling] Comparison of R2 proposed IM classes and IFA011 Hi- I'm trying to understand how the IM model proposed for R2 compares to IFA011. When I look on this wiki: https://wiki.onap.org/display/DW/Resource+IM+Discussion+Based+on+IFA011 it looks like the IFA011 model was reviewed and decisions were made. For example, in the VNFD, one decision was to rename vnfProductName to Name. It is marked as "AGREED". However, when I look at the R2 class VNFD, the attribute is still called "vnfProductName". Should I be ignoring the decisions made on the wiki above? Is there another place that shows the differences between the R2 classes and IFA011. Thanks for your help, Jessie
_______________________________________________ onap-discuss mailing list [email protected] https://lists.onap.org/mailman/listinfo/onap-discuss
