[Smwg] RE: Abstract service request telecon notes

Tue Jul 28 13:11:42 UTC 2015

Hi Erik, Holger,

First sorry for not attending last telecons (or badly attending) but I just could not manage that. I'll try to get better on that, however tomorrow I will not be able to support the telecon... (so much for my trying, huh?).

Anyway, for the small discussion below, some of my thoughts:

For the DLR we are not yet "service" oriented, which means we do not differentiate between a pass and the service which is behind it (or with it, or within, or whatever). Also as you know we do not have all the great things of service agreements, configuration profiles, etc... Now, what we have in principle is just station allocation and ... that's it. All further details are happening "somehow". Which does not mean, we would not like to go for full integration and automation someday, where more precise information (like service type, on/off times, etc... ergo service scheduling) may be required.

In general I could agree with Holger having more or less two different areas (station allocation and actual service scheduling), whereas I also thought (until now at least) this was intended by SM to have service scheduling und through that have more or less indirectly station allocation.

We could say than we have (in narrow terms of service and station requests only) following topics:

1.       Allocation / service planning

2.       Station allocation/booking in general

3.       Service scheduling in detail

I do not see (directly) any reasons why we may not split point 2 and 3 from each other. This would actually mean, that the agencies like DLR which up to now use rather rudimentary scheduling, would be probably happy to use just point 2 request type.

This does not change anything on other uses for our "mother-of-all-requests" (short: MOAR), where we may use it for requesting several different things (like service agreement, or later on accounting). And it would be nice, if we manage to have common constraints class for all of these types - I think we can do this, because 99% of constraints we will have, cover timings, service types, apertures, spacecraft, etc. All these things are important (and common!) for planning, allocation, services, accounting, service agreement.

Regarding the question if constraints should be composed or referenced - at the first shot, as we speak of "requests", I would also agree, that they should be composed (as being special service, special case related). On the other side I may also imagine the case where specific constraints being already predefined (whatever way) and then just referenced. In such a case however we need the composition possibility anyway.

Also is pity I can't participate tomorrow, as Erik promises to go more into details and be less abstract. I hope you'll make some notes of the meeting, so that I can follow up.

I hope I was not too chaotic, so my input has some value ;-).

Best Regards
Marcin

From: Barkley, Erik J (3970) [mailto:erik.j.barkley at jpl.nasa.gov]
Sent: Freitag, 3. Juli 2015 02:40
To: Holger.Dreihahn at esa.int
Cc: Colin.Haddow at esa.int; Chamoun, Jean-Pierre (GSFC-458.0)[Affiliate] (jean-pierre.chamoun at nasa.gov); Gnat, Marcin; CCSDS Service Mgmt WG; Unal, Martin P (9020-Affiliate)
Subject: RE: Abstract service request telecon notes

Holger,

I recognize more or less the process you have described and there are definite similarities to what I am most familiar with in terms of business process model, lifecycle, for delivery of TT+C services for the DSN.    But there may be other organizations that don't quite do things the same way or perhaps are using a request in sense that does not have the all of the supporting details (such as service agreement, configuration profile, event sequence) etc. in place.  For example, assuming we are wildly successful (why not?) it may be that someday we have a normative recommendation includes  a request for creation of a service agreement.  It may be that there are constraints expressed for each service on an independent basis or at least in some relatively de-coupled manner.

Part of what we are attempting to do here is understand what are the components needed for a service request in the abstract and so this is probably best viewed as an M-2 model.   The idea is to understand  what and where are the functional points of articulations for requests in general.  Its kind of functional resources for service request.  At the M-2 level I tend to be satisfied that there is, in the abstract, a constraints function.  This then is an essential extensibility point  that can be subject to restrictions such as you indicated below (ie, constraints only on the entire package of services).

I think we will know more at our next discussion on this when we see the planning request vs (a more exact DDOR flavored) service package request being compared and contrasted.  In both cases there will likely be some sort of constraints, some sort of service configuration function, some sort of ability to reference other information, etc - these are the  items of primary interest at the moment.

Best regards,

-Erik

From: Holger.Dreihahn at esa.int<mailto:Holger.Dreihahn at esa.int> [mailto:Holger.Dreihahn at esa.int]
Sent: Wednesday, July 01, 2015 2:43 AM
To: Barkley, Erik J (3970)
Cc: Colin.Haddow at esa.int<mailto:Colin.Haddow at esa.int>; Chamoun, Jean-Pierre (GSFC-458.0)[Affiliate] (jean-pierre.chamoun at nasa.gov<mailto:jean-pierre.chamoun at nasa.gov>); Marcin.Gnat at dlr.de<mailto:Marcin.Gnat at dlr.de>; CCSDS Service Mgmt WG; Unal, Martin P (9020-Affiliate)
Subject: Re: Abstract service request telecon notes

Hi Erik,
Thanks for the opportunity to join the discussion. Having a second thought about the request and the discussed option to attach constraints to the provided services I was considering the following:

- for the request itself constraints make sense to express flexibility, something like 'I need a an 8-10h pass starting earliest at 12h, latest start at 14h.
- for the services within the pass I am not that convinced (anymore):
        \ not all required details might be known at request time
        \ according to my experience missions want to prescribe an exact timing when services (uplink, ranging, doppler measurement etc.) are provided within a pass

In our network we now at pass request time the set of required services, the station capability in terms of service provisioning and perform the booking (station allocation plan). However, at that point in time some details for the service provisioning within the pass are not known. Typically at a later stage missions provide (often by providing mission specific events) the details, which allows the production of executable station schedule, which actually refines the station allocation plan into a more detailed schedule which logically contains the service provision times (and parameterisation) of the provided services within a pass. In addition to that there may be the need to update the pass allocation timing at a later stage based on updated orbital predictions.

I would expect that scenario to be typical for network: Resources (station and others) are booked quite in advance. At this time often not all information (e.g. orbital predictions) is available at the finally required precision to actually schedule station services correctly. If that is true, shouldn't we reflect that in the abstract service request, i.e. simplify it to what's needed. In our network the equivalent of the service agreement captures the rules when services are provided within a pass, often this is done by referring mission specific events. Now I do not know exactly, but I would expect the sequence of events to provide a similar sort of functionality.

As such it looks to me we are mixing up two problems: Station (resources) allocation and service scheduling within the allocated passes. Is that necessarily a good idea? I think this is a fundamental point we need to discuss. According to a discussion with Martin the resource allocation request need the outlined flexibility to facilitate better resource exploitation. Services within a pass have to be provided according to precise timings with agreed and validated (parameterised)  setup.
The resource allocation is typically an input to mission planning, which results later in the details required to provide the service within the allocated resource booking.

For the constraints of the request itself I would rate composition (black diamond) as the right way of doing it. For service specific constraints (or details, rules or what it will be) I tend to agree that those should be references to, ja, do not know exactly but some (static) configuration like the rules captured by the service agreement.

Kind regards,
Holger

Holger Dreihahn
European Spacecraft Operations Centre | European Space Agency | S-431
+49 6151 90 2233 | http://www.esa.int/esoc

From:        "Barkley, Erik J (3970)" <erik.j.barkley at jpl.nasa.gov<mailto:erik.j.barkley at jpl.nasa.gov>>
To:        "Chamoun, Jean-Pierre (GSFC-458.0)[Affiliate] (jean-pierre.chamoun at nasa.gov<mailto:jean-pierre.chamoun at nasa.gov>)" <jean-pierre.chamoun at nasa.gov<mailto:jean-pierre.chamoun at nasa.gov>>, "Marcin.Gnat at dlr.de<mailto:Marcin.Gnat at dlr.de>" <Marcin.Gnat at dlr.de<mailto:Marcin.Gnat at dlr.de>>, "Colin.Haddow at esa.int<mailto:Colin.Haddow at esa.int>" <Colin.Haddow at esa.int<mailto:Colin.Haddow at esa.int>>, "Holger.Dreihahn at esa.int<mailto:Holger.Dreihahn at esa.int>" <Holger.Dreihahn at esa.int<mailto:Holger.Dreihahn at esa.int>>,
Cc:        CCSDS Service Mgmt WG <smwg at mailman.ccsds.org<mailto:smwg at mailman.ccsds.org>>
Date:        01/07/2015 03:38
Subject:        Abstract service request telecon notes
________________________________

Colin, Marcin, JP, Holger,

Attached are the notes from our June 30 splinter session teleconference on engineering an abstract service request. The notes have also been uploaded to the CWE. Corrections and/or comments appreciated.

Best regards,

-Erik[attachment "2015-06-30-AbstractSvcReq-TeleconferenceNotes.docx" deleted by Holger Dreihahn/esoc/ESA]

This message and any attachments are intended for the use of the addressee or addressees only.

The unauthorised disclosure, use, dissemination or copying (either in whole or in part) of its

content is not permitted.

If you received this message in error, please notify the sender and delete it from your system.

Emails can be altered and their integrity cannot be guaranteed by the sender.

Please consider the environment before printing this email.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ccsds.org/pipermail/smwg/attachments/20150728/d33d1254/attachment.html>