Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [papyrus-ic] Support for Papyrus UML added to Hawk model indexer 
Hi Johan,

Disclaimer: I am not part of the IncQuery-D project, but I was a Research Associate in the MONDO project and had some experience with plain IncQuery.

IncQuery(-D) and Hawk are aimed at different types of queries. IncQuery and IncQuery-D focus on incremental queries over specific models that they have loaded in memory, by turning the pattern matching query into some representation (a RETE-style network for IncQuery, IIRC) that can propagate changes in the model back to the results (the pattern matches). IncQuery-D is a distributed version that can handle bigger models and runs on Hadoop clusters, I believe.

In contrast, Hawk is designed to quickly answer one-shot queries about large collections of interconnected models  (like all the models in your Git/SVN repositories or your workspace) at once. These are mirrored on a per-file basis into a graph DB, without requiring a complex deployment or changing the way you work with your models normally.  Hawk has some extra features (like pre-computed derived features or index-based lookups) which I believe are not part of IncQuery.

We are considering introducing horizontal scaling / some form of incremental querying in the future, but we are tackling other priorities at the moment like widening the types of models we support (hence the addition of UML) and improving the user experience.

Hope that helped - feel free to ask more!

Kind regards,
Antonio

--
Antonio Garcia-Dominguez
Lecturer in Computer Science
School of Engineering & Applied Science, Aston University
Aston Triangle, Birmingham, B4 7ET
Room: MB211Q


________________________________________
From: papyrus-ic-bounces@xxxxxxxxxxx [papyrus-ic-bounces@xxxxxxxxxxx] on behalf of Johan Van Noten [johan.vannoten@xxxxxxxxxxxxxxx]
Sent: 19 January 2018 19:29
To: papyrus-ic discussions
Subject: Re: [papyrus-ic] Support for Papyrus UML added to Hawk model indexer

Hi Antonio,

I see that Hawk is part of the Mondo-project (https://emea01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.mondo-project.org%2Ftechnologies&data=02%7C01%7Ca.garcia-dominguez%40aston.ac.uk%7C7faa89955af545c9987308d55f730b34%7Ca085950c4c2544d5945ab852fa44a221%7C0%7C0%7C636519870074512082&sdata=4%2B1w%2FK6JQijmM6j878WNE026bBxOZZjbro5cfgmfF2s%3D&reserved=0).
The description that you give makes me think about my experiences with Viatra/IncQuery, coming from the same project.
Could you briefly describe the main relationships/differences between both?
Looking at the wiki on https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fmondo-project%2Fmondo-hawk%2Fwiki&data=02%7C01%7Ca.garcia-dominguez%40aston.ac.uk%7C7faa89955af545c9987308d55f730b34%7Ca085950c4c2544d5945ab852fa44a221%7C0%7C0%7C636519870074512082&sdata=rAJGZXpNzPMXihLw70fWP3yA4tgGgzJcMgGCL1C%2FrP8%3D&reserved=0 doesn't fully clarify this for me.

Johan

-----Oorspronkelijk bericht-----
Van: papyrus-ic-bounces@xxxxxxxxxxx [mailto:papyrus-ic-bounces@xxxxxxxxxxx] Namens Garcia-Dominguez, Antonio
Verzonden: donderdag 18 januari 2018 18:29
Aan: papyrus-ic@xxxxxxxxxxx
Onderwerp: [papyrus-ic] Support for Papyrus UML added to Hawk model indexer

Hello all,

We have recently added support to index UML models to our Hawk tool:

https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fmondo-project%2Fmondo-hawk%2Fwiki%2FPapyrus-UML-support-in-Hawk&data=02%7C01%7Ca.garcia-dominguez%40aston.ac.uk%7C7faa89955af545c9987308d55f730b34%7Ca085950c4c2544d5945ab852fa44a221%7C0%7C0%7C636519870074512082&sdata=kfo8qn4buIOoQw0yNo1nLABzaZATuLok0%2B8oSOhME6g%3D&reserved=0

The tool can automatically watch all the models in the workspace (also folders and SVN/Git repositories) and mirror them into a graph database (currently Neo4j or OrientDB) for faster and more convenient querying. We also have provisions for convenient querying of models that use profiles, and for models that use the predefined UML libraries as well.

The wiki article has a full tutorial on how to use it with UML, and a few sample queries on a toy model. We'd love to know your thoughts on this, especially if we are missing something important in terms of supporting UML in industrial settings.

By the way, can we post this to the wider community in mdt-papyrus.dev? We are not sure if it would fit in there.

Looking forward to your feedback!
Antonio

--
Antonio Garcia-Dominguez
Lecturer in Computer Science
School of Engineering & Applied Science, Aston University Aston Triangle, Birmingham, B4 7ET
Room: MB211Q

_______________________________________________
papyrus-ic mailing list
papyrus-ic@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdev.eclipse.org%2Fmailman%2Flistinfo%2Fpapyrus-ic&data=02%7C01%7Ca.garcia-dominguez%40aston.ac.uk%7C7faa89955af545c9987308d55f730b34%7Ca085950c4c2544d5945ab852fa44a221%7C0%7C0%7C636519870074512082&sdata=y585Tmwo5flGYQ7U7wUb7krzdjJ81uO0KOjuAxcygRc%3D&reserved=0
_______________________________________________
papyrus-ic mailing list
papyrus-ic@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdev.eclipse.org%2Fmailman%2Flistinfo%2Fpapyrus-ic&data=02%7C01%7Ca.garcia-dominguez%40aston.ac.uk%7C7faa89955af545c9987308d55f730b34%7Ca085950c4c2544d5945ab852fa44a221%7C0%7C0%7C636519870074512082&sdata=y585Tmwo5flGYQ7U7wUb7krzdjJ81uO0KOjuAxcygRc%3D&reserved=0

Back to the top