[NMOP] Re: Document shepherd review of draft-ietf-nmop-simap-concept-08

[Reshad Rahman <[e-mail]>]

 Hi Olga,
I am good with the changes below.
Regards,Reshad.
    On Wednesday, March 18, 2026 at 03:48:11 AM EDT, Olga Havel <olga.havel=[e-mail]> wrote:  
 
 Dear all,

There is a new version of the SIMAP concept draft with the changes based on Reshad's feedback.

https://datatracker.ietf.org/doc/draft-ietf-nmop-simap-concept/09/

The github pull request is
https://github.com/ietf-wg-nmop/draft-ietf-nmop-digital-map-concept/pull/172

Best Regards,
Olga

-----Original Message-----
From: Olga Havel 
Sent: Monday, March 16, 2026 12:31 PM
To: 'Reshad Rahman' <[e-mail]>; '[e-mail]' <[e-mail]>; 'Nmop' <[e-mail]>
Subject: RE: Document shepherd review of draft-ietf-nmop-simap-concept-08

Hi Reshad,

Thanks again for your review. What about the following changes for 1, 2 and 4. I hope 3 and 5 only require clarification, but let me know if you suggest any changes. I will submit the new version as soon as I get your feedback to the proposal.

1. Abstact - Digital Map:

    [AS-IS]
      The SIMAP was previously known as Digital Map.
    [TO-BE]
    The SIMAP was previously known as Digital Map. 
    SIMAP evolves the earlier ‘Digital Map’ concept by making explicit the ties between service and infrastructure layers, clarifying expected outcomes for operations and automation, and addressing ambiguity associated with the term ‘digital.’

2. Introduction - Topological Reasoning:

    [AS-IS]
    The core topological properties are the essential attributes associated with these entities to enable topological reasoning (e.g., identity, topology type, entity role in topology, directionality, cardinality, and cost/weight).
    [TO-BE]
    The core topological properties are the essential attributes associated with these entities (e.g., identity, topology type, entity role in topology,  directionality, cardinality, and cost/weight), enabling analysis of how the network structure affects services and operations. For example, topological reasoning can answer questions such as: 'If link X fails, what services are impacted?' or 'What is the full end-to-end data path of the service flow?'.

3. Introduction - not to change a topology layer on the fly:

    Clarification for: But I don't get the "not to change a topology layer on the fly" part. If there's a change in the physical network, won't the SIMAP be updated on the fly to represent the new state of the physical network.

    [Olga] It will be changed, but the SIMAP server will change it via discovery and it will not be changed by the SIMAP client via write operation. We were asked to explicitly add it so that it is clear that the write will only be used for simulations / what-if. Please let me know if this clarifies or if you want any change in regards to this comment.

4. Introduction - Real versus Online versus offline simulation.

    AS-IS:    Both real network, online-simulation and offline-simulation APIs can be built on the same data model.
    TO-BE:  Both real network, online simulation, and offline simulation APIs can be built on the same data model. The real network API reflects actual changes in the topology as reported by the SIMAP server. Online simulation applies hypothetical changes to the current live model to assess immediate impacts (e.g., if link X fails, what services are disrupted), without altering the real network. Offline simulation applies hypothetical changes to a saved or alternate model, useful for planning, training, or evaluating changes before deployment.”

5. Terminology (CoS versus type of service)
    I consider Class of service as video, voice, data - flows inside the service for a specific class. Type of service would be L2VPN, L3VPN, 
    Copilot tells me the following:
        Class of Service (CoS): In networking, this usually refers to traffic categories used for QoS (e.g., video, voice, data). 
        Type of Service (ToS): Historically an IP header field, but in IETF usage it can also mean service types like L3VPN, L2VPN, etc. 

    Please let me know if this clarification is enough or if you think we need to make any change.


Best Regards,
Olga

-----Original Message-----
From: Reshad Rahman <[e-mail]> 
Sent: Monday, March 9, 2026 11:53 PM
To: [e-mail]; Nmop <[e-mail]>
Subject: Document shepherd review of draft-ietf-nmop-simap-concept-08

Authors,
Thank you for addressing the WGLC comments to closure.
My comments are below, I have strived not to reopen any previously closed issues so apologies in advance if I inadvertently did...
Regards,Reshad.
Abstract
   The SIMAP was previously known as Digital Map.- I didn't see anywhere in the document the reason for the name change and how SIMAP is different from Digital Map.Suggestion: in the abstract or introduction please add some text. e.g. SIMAP evolves the earlier “Digital Map” concept to emphasize explicit ties between service and infrastructure layers and to clarify expected outcomes for operations and automation.

1. Introduction
   The core topological properties are the essential   attributes associated with these entities to enable topological   reasoning (e.g., identity, topology type, entity role in topology,   directionality, cardinality, and cost/weight).- What is "topological reasoning"? It looks this was added to resolve one of the many WGLC comments, I don't think it's a term which is commonly used in IETF documents. Could you explain further what it means and/or add an example e.g. "if link X fails, what services are impacted?".
   The SIMAP also provides write operations with the same set of APIs,   not to change a topology layer on the fly as a northbound interface   from the controller, but for both online and offline simulations,   before applying the changes to the network via the normal controller   operations.- I am a bit confused by this paragraph. My understanding is that the write APIs can be used for "what if" scenarios e.g "what is the impact of taking that node or link down", that's what the "online and offline simulations" is referring to.  But I don't get the "not to change a topology layer on the fly" part. If there's a change in the physical network, won't the SIMAP be updated on the fly to represent the new state of the physical network.
   Both real network, online-simulation and offline-simulation APIs can   be built on the same data model.- Can you clarify the differences between online and offline simulations? The previous version mentioned offline simulation but online simulation was added in 08. Also, what's the difference between real network and online simulations?
2. Terminology 
   The application/flow layer represents the view of Service data   flows for different classes of service - video, voice and data   traffiC. - By "classes of service", do you mean "types of services"? Class of service is QOSish to me.  Nit: traffiC -> traffic
-- 
Nmop mailing list -- [e-mail]
To unsubscribe send an email to [e-mail]