Validation of Code_Saturne

Questions and remarks about code_saturne usage
Forum rules
Please read the forum usage recommendations before posting.
Post Reply
César Vecchio

Validation of Code_Saturne

Post by César Vecchio »

Hello all,
I'm writing a technical report for my boss about some simulations I made with C_S, and I would like to know if C_S has some kind of certification like ISO 9001 or alikes. I know Code_Aster has, and since the owning company is the same I suppose the same kind of verifications apply, but I don't want to write something that may not be true.
It's not mandatory for me to include this, but since this results will be shown in the company's brochures, naming a certified and validated software does certainly boost the image of the company :D. Would you mind telling me where can I find more information about this, or describe the validation/certification process (just some lines, nothing too detailed)?
Thank you in advance and happy New Year.
 
César
James McNaughton

Re: Validation of Code_Saturne

Post by James McNaughton »

At University of Manchester our twiki page shows a number of test cases that are used for validation. Not sure if it's what you need but might be worth a look http://cfd.mace.manchester.ac.uk/twiki/bin/view/Saturne/TestCases
James
Yvan Fournier

Re: Validation of Code_Saturne

Post by Yvan Fournier »

Hello,
As James indicated, at least part of the validation is public, and the code is developped under quality assurance. I do not know which version you use, but version 1.3.3 is an officially validated version. The same holds true for 2.0.0 (and corrective releases 2.0.1 and others which may follow), but intermediate versions 2.0-beta1/2 or 2.0-rc1/2 are not officially validated versions (2.0.0-rc2 results from the first validation pass corrections, so it is close, but still an intermediate version).
Best regards,
  Yvan
César Vecchio

Re: Validation of Code_Saturne

Post by César Vecchio »

Great, I think that should be enough... too bad I use the 2.0-beta2. Thank you both!
Nima Maftoon

Re: Validation of Code_Saturne

Post by Nima Maftoon »

Hello,
I have similar question. I know that Code_Aster is verified extensively which is reflected in their V-series of manuals. They have some access restriction to some manuals on the web which is understandable.
I am sure in the same spirit, Code_Saturne is dependable. Personally I am attracted to Code_Saturne because I think there should be some kind of verification. However we need to know whether Code_Saturne has (or will have) a verification manual. Furthermore users should know extent of validity of each capability of the code. What is your plan to publish those data?
Thanks
Nima
Yvan Fournier

Re: Validation of Code_Saturne

Post by Yvan Fournier »

Hello,
I'll let David Monfort complete this with more details, but the validation manual for Code_Saturne 2.0 was released a few days ago as an internal EDF document.
Test cases should also be published on the University of Manchester Twiki (http://cfd.mace.manchester.ac.uk/twiki/bin/view/Saturne/TestCases), but they have not been released yet. To my knowledge, they should be public soon, but as validation requirements have changed/increased (compared to  simple publication of results), which has increased the delays, I'll wait for others to confirm this.
Also, the scope of the official validation is the core solver (turbulent incompressible flows with or without heat transfer). Specific physics such as combustion have their own, separate validation, as they are managed mainly by other projects, so I can't speak for those.
Note also that version 2.0 is a validated version, while 2.1 is a development release, which has been extensively tested, but is not the object of a full official validation (basically, starting with version 2.0, you can see this as standard vs. long-term support variants of Linux distributions : x.0 versions have a full, "official" validation, while x.1, x.2, and x.3 versions are intermediate release, with new features, but less systematic testing.
Alexandre and David are also working on a more automated validation/regression testing framework, so intermediate versions should be as "safe" as long-term versions when this is ready, but this is still work in progress.
Hoping this helps you wait for the "published" validation.
Best regards,  
Yvan
Fanny.DM
Posts: 6
Joined: Fri Mar 27, 2020 8:35 am

Re: Validation of Code_Saturne

Post by Fanny.DM »

Hello,

Sorry for reopening an old post, but I'm interested in the validation of Code-Saturne and Syrthes for thermal applications, and the link you gave here aren't working anymore.

I'm using version 3.0.1 (old cluster) and version 7.0 (new cluster) of Code-Saturne for thermal nuclear studies and I have to justify the validation and qualification of both codes (with a priority for 3.0.1 version). Is there a public document assessing of the completion of the validation process of Code-Saturne for this type of studies?

Thank you for your help,

Fanny DM.
Yvan Fournier
Posts: 4070
Joined: Mon Feb 20, 2012 3:25 pm

Re: Validation of Code_Saturne

Post by Yvan Fournier »

Hello,

Unfortunately, we are not authorized to make the validation public, and that includes the validation reports.

We hope to make a subset of our validation suite public in the near future, as this would provide good examples for use, but this requires curating the test cases to see which ones become public or not bases on various criteria, and takes time for an already busy team).

Best regards,

Yvan Fournier
Post Reply