RE: [tosca] Proposal for requirement "occurrences"

[Chris Lauwers <lauwers@ubicity.com>]

In a âdesignâ, everything is a template. This means that in a topology template graph, the edges between the node templates are relationship templates. The first step in deploying a service (the step Iâd like to refer to as âprovisioningâ) is to create representations based on the templates: node representations are created from node templates, and relationships are created from relationship templates.

 

Unfortunately, this relationship template concept is far from obvious from the current spec:

 

• The âexplicitâ definition of relationship templates (at the topology template level) is broken and as a result almost unusable
• The relationship templates that are actually usable are part of the ârequirement assignmentâ grammar and in almost all cases just consist of a ârelationship typeâ keyword (since very few relationships actually defined properties or interface operation implementations).

 

Yes, we should discuss this in the ad-hoc and get agreement on the basic concepts before we talk about specific grammar. Unfortunately I have a conflict this week so weâll have to postpone until next week.

 

Thanks,

 

Chris

 

 

From: Tal Liron <tliron@redhat.com>
Sent: Monday, February 7, 2022 9:32 AM
To: Chris Lauwers <lauwers@ubicity.com>
Cc: Calin Curescu <calin.curescu@ericsson.com>; tosca@lists.oasis-open.org
Subject: Re: [tosca] Proposal for requirement "occurrences"

 

On Mon, Feb 7, 2022 at 11:18 AM Chris Lauwers <lauwers@ubicity.com> wrote:

 

• If youâre visualizing a design (e.g. the Cloudnet demo) then youâre visualizing a topology template graph that consists of node templates
• If youâre visualizing a service instance (that is ready to be activated or has been activated) then youâre visualizing a topology representation graphs that consists of node representations.

Let's focus on the first bullet point. What you call a "topology template graph" is what I've called "design topology", which you and Calin have insisted is not a real thing in TOSCA because relationships only exist between node representations. So, what are these edges that we're seeing? I'm trying to get to the bottom of it. I want us to be very clear on what that actually is and how it relates to the second bullet point.

 

I suggest we have a discussion in the ad hoc about this because email is not getting us very far.