[Paraview] Concept: A world without the Apply button

Wed Jun 8 09:11:47 EDT 2011

[Stephane and I have been considering a proposal to add some degree of caching into the pipeline which would handle exactly the kind of situation where certain arrays could be passed through without the need for portions of the pipeline to reexecute.] (I'd like to see RequestData broken into smaller RequestScalar/RequestMesh type requests too, by the way)

However, your description of changes (which I basically approve of, subject to solving the issue of filters changing type based on input type the Dave pointed out) hints at a more contract based approach to updating the pipeline and if we're going to go down this path, I think we need to completely reconsider how extents are handled. Piece updates and extent based updates are (in my opinion) a bit shoddy and temperamental and the RequestUpdataExtent doesn't well (or at all) handle geometric extents (as opposed to structured extents or pieces). Also, the streaming improvements that Dave has been adding look marvellous, but appear to be somewhat orthogonal to the rest of the update mechanism and I'd it be nice to see these changes as part of a broader look at improving the handling of load balancing and extent handling within the entire executive framework -a more unified way.

I think I got a bit off topic there. Sorry

JB

From: paraview-bounces at paraview.org [mailto:paraview-bounces at paraview.org] On Behalf Of Utkarsh Ayachit
Sent: 08 June 2011 14:50
To: Stephane PLOIX
Cc: paraview at paraview.org
Subject: Re: [Paraview] Concept: A world without the Apply button

Stephane,

Yes, that'd indeed be a path forward, but like you alluded to, it would need a planning of its own especially adding support to the executive, being smart about when to re-execute, and even possibly, cache arrays.

Utkarsh
On Wed, Jun 8, 2011 at 8:33 AM, Stephane PLOIX <stephane.ploix at edf.fr<mailto:stephane.ploix at edf.fr>> wrote:

Hi Utkarsh,

If we want to push the demand-driven logic to the end, choosing the arrays to load could be implicit on the readers : they will depend on the arrays needed down the pipeline.
The obvious drawback of this approach is that having several arrays availbale for the representation to color by is very handy, instead of re-executing the whole pipeline each time you change the array to color by, but this demand-driven logic could help when memory is limited.

Best,
Stephane

 [cid:image001.gif at 01CC25EC.DDE2A730]

Stephane PLOIX
Pilote Opérationnel - Visualisation scientifique
EDF - R&D
SINETICS
1, Av du Général de Gaulle
92140 Clamart

stephane.ploix at edf.fr<mailto:stephane.ploix at edf.fr>
Tél. : +33 (0) 1 47 65 51 10

 [cid:image002.gif at 01CC25EC.DDE2A730]

Un geste simple pour l'environnement, n'imprimez ce message que si vous en avez l'utilité.

utkarsh.ayachit at kitware.com<mailto:utkarsh.ayachit at kitware.com>
Envoyé par : paraview-bounces at paraview.org<mailto:paraview-bounces at paraview.org>

07/06/2011 19:44

A

samuelkey at bresnan.net<mailto:samuelkey at bresnan.net>

cc

paraview at paraview.org<mailto:paraview at paraview.org>

Objet

Re: [Paraview] Concept: A world without the Apply button

Sam,

Please note that this proposal does not preclude the user from
choosing which array to load from the reader's properties panel. That
indeed will be supported. This proposal in fact allows you go back the
reader change it's array selection, and then go the properties panel
for a contour filter connected downstream (for example) and choose the
array to contour by as the newly selected array without having to
execute any of the reader/intermediate filters until you have setup
the full pipeline. After which, you hit "Execute". That's when the
reader will actually read the data. Currently, before you can choose
the array to contour with, you have to hit "Apply" so that the reader
will produce the newly selected array. Then alone will it be reflected
on the contour panel.

Also, changing parameters on filters will have no immediate effect per
say. The new proposal is not simply "auto-apply". You have to hit
"Execute" before the filters will be executed, so you can tweak
parameters to your hearts contents before triggering any execution
(similar to the Apply mechanism).

Utkarsh

On Tue, Jun 7, 2011 at 1:31 PM, Samuel Key <samuelkey at bresnan.net<mailto:samuelkey at bresnan.net>> wrote:
> All--
>
> Technical reasons aside, my own simulation efforts (software that relies on
> ParaView for post-processing) span a great variety of solid and structural
> mechanics applications. The software is not a single-formulation,
> single-solution, one-off problem solver.
>
> Put briefly, I find the "Apply" button exceptionally useful.
>
> The EnSight-formatted results files produced contain a surfeit (excess) of
> simulation results for any one application. I use the "Apply" button on
> loading results files for several purposes:
>
> 1) Load only results that are most useful to the current problem or
> application.
>
> 2) Limit the results I examine to those relevant to the phase st hand:
>     i) Phase 1. Are the results OK (BC's OK? Energy Balance OK? Loading OK?)
>    ii) Phase 2. Which result describes the response most effectively?
>   iii) Phase 3. Presentation animations and figures for communicating.
>
> 3) Frequent tuning of various filters to refine the display of the results
> being studied.
>
> 4) Last but not least, limit the results being examined at any one session
> to avoid the MS O/S crashing PV because of the MS O/S limits on resources
> usable.
>
> Thanks for the opportunity to express my support for ParaView and its
> exceptional versatility and effectiveness as a post-processing tool.
>
> Sam Key
>
> On 6/7/2011 8:56 AM, Moreland, Kenneth wrote:
>>
>> One thing that strikes me about this proposal, is that it relies a lot on
>> VTK readers/filters providing the necessary meta data, particularly about
>> fields.  For example, the elevation filter would have to report in its
>> meta data that it will be adding a scalar field named Elevation.  That
>> sounds easy enough to implement, but would have to be independently added
>> for all readers/filters.  This would be especially problematic for
>> user-provided plugins.
>>
>> Is there a way to mitigate the issues with the meta data mismatching what
>> is actually generated?  Perhaps there could be a check to make sure that
>> the data generated matches the meta data previously reported and issue a
>> warning if there are any mismatches.  This would not solve the problem per
>> se, but would quickly alert developers to problems and make it easier to
>> debug.
>>
>> -Ken
>>
>>    ****      Kenneth Moreland
>>     ***      Sandia National Laboratories
>> ***********
>> *** *** ***  email: kmorel at sandia.gov<mailto:kmorel at sandia.gov>
>> **  ***  **  phone: (505) 844-8919<tel:%28505%29%20844-8919>
>>     ***      web:   http://www.cs.unm.edu/~kmorel
>>
>>
>>
>>
>> On 6/1/11 1:30 PM, "Utkarsh Ayachit"<utkarsh.ayachit at kitware.com<mailto:utkarsh.ayachit at kitware.com>>  wrote:
>>
>>> Folks,
>>>
>>> I have been investigating the ability to create pipelines in ParaView
>>> without having to hit Apply at every single stage in the pipeline.
>>> This would make it much easier to deal with really large data in a
>>> usage scenario where the scientist is fairly familiar with the
>>> visualization pipeline he's interested in setting up.
>>>
>>> Following the recent changes to the ParaView ServerManager, the
>>> ServerManager level requirement that filters need to be updated before
>>> connecting has disappeared. However, we still need to "Apply" to get
>>> updated information about arrays etc.
>>>
>>> After a couple on discussions with Berk and Dave, I've consolidated
>>> the thoughts on a Wiki page. If any one has any thoughts to share, it
>>> would be great.
>>>
>>>
>>> http://www.paraview.org/ParaView3/index.php/No-Apply_Mechanism_for_Creatin
>>> g_Pipelines
>>>
>>> Note that this is merely a concept. We have no plans of implementing
>>> it in near future (3.12/4.0).
>>>
>>> Utkarsh
>>> _______________________________________________
>>> Powered by www.kitware.com<http://www.kitware.com>
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Please keep messages on-topic and check the ParaView Wiki at:
>>> http://paraview.org/Wiki/ParaView
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://www.paraview.org/mailman/listinfo/paraview
>>>
>>
>> _______________________________________________
>> Powered by www.kitware.com<http://www.kitware.com>
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Please keep messages on-topic and check the ParaView Wiki at:
>> http://paraview.org/Wiki/ParaView
>>
>> Follow this link to subscribe/unsubscribe:
>> http://www.paraview.org/mailman/listinfo/paraview
>>
>
_______________________________________________
Powered by www.kitware.com<http://www.kitware.com>

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview

Ce message et toutes les pièces jointes (ci-après le 'Message') sont établis à l'intention exclusive des destinataires et les informations qui y figurent sont strictement confidentielles. Toute utilisation de ce Message non conforme à sa destination, toute diffusion ou toute publication totale ou partielle, est interdite sauf autorisation expresse.

Si vous n'êtes pas le destinataire de ce Message, il vous est interdit de le copier, de le faire suivre, de le divulguer ou d'en utiliser tout ou partie. Si vous avez reçu ce Message par erreur, merci de le supprimer de votre système, ainsi que toutes ses copies, et de n'en garder aucune trace sur quelque support que ce soit. Nous vous remercions également d'en avertir immédiatement l'expéditeur par retour du message.

Il est impossible de garantir que les communications par messagerie électronique arrivent en temps utile, sont sécurisées ou dénuées de toute erreur ou virus.
____________________________________________________

This message and any attachments (the 'Message') are intended solely for the addressees. The information contained in this Message is confidential. Any use of information contained in this Message not in accord with its purpose, any dissemination or disclosure, either whole or partial, is prohibited except formal approval.

If you are not the addressee, you may not copy, forward, disclose or use any part of it. If you have received this message in error, please delete it and all copies from your system and notify the sender immediately by return message.

E-mail communication cannot be guaranteed to be timely secure, error or virus-free.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20110608/9a03fbf5/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.gif
Type: image/gif
Size: 1816 bytes
Desc: image001.gif
URL: <http://www.paraview.org/pipermail/paraview/attachments/20110608/9a03fbf5/attachment-0002.gif>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.gif
Type: image/gif
Size: 1151 bytes
Desc: image002.gif
URL: <http://www.paraview.org/pipermail/paraview/attachments/20110608/9a03fbf5/attachment-0003.gif>