[petsc-users] Performance differences between PETSc4py and PETSc on C

[Jed Brown]

Nidish <nb25 at rice.edu> writes:

> Does that mean one can choose to develop code using the petc4py wrappers 
> without having to sacrifice any performance?

Roughly, yes.  Though assembly of residuals and Jacobians is still your business and its performance will vary greatly with the class of method and your implementation.

> Apologies if this question is too basic, I'm just trying to understand 
> what I would be sacrificing if I chose to completely write my 
> application on Python.
>
> Nidish
>
> On 8/13/20 8:43 PM, Matthew Knepley wrote:
>> On Thu, Aug 13, 2020 at 9:42 PM Nidish <nb25 at rice.edu 
>> <mailto:nb25 at rice.edu>> wrote:
>>
>>     Hello,
>>
>>     I'm wondering if any performance studies have been conducted between
>>     codes written using PETSc on C versus Python implementations using
>>     PETSc4py. Other than this, I'd really appreciate it if someone can
>>     give
>>     perspectives on the drawbacks/advantages on opting for either for
>>     code
>>     development.
>>
>>
>> No PETSc code is in Python, so there is no difference as long as you 
>> do not call PETSc
>> function millions of times from Python. It should always be possible 
>> to operate at the right
>> granularity.
>>
>>   Thanks,
>>
>>      Matt
>>
>>     Thank you,
>>     Nidish
>>
>>
>>
>> -- 
>> What most experimenters take for granted before they begin their 
>> experiments is infinitely more interesting than any results to which 
>> their experiments lead.
>> -- Norbert Wiener
>>
>> https://www.cse.buffalo.edu/~knepley/ 
>> <http://www.cse.buffalo.edu/~knepley/>
> -- 
> Nidish