[QE-users] [neb.x] How to lessen I/O?

[Takahiro Chiba]

Thank you Giannozzi, I will watch the issue at gitlab.

---Sender---
Takahiro Chiba
Undergraduate at Hokkaido University
Expected graduation date: Mar. 2021
Expected to be a graduate student at HU from Apr. 2021
takahiro_chiba at eis.hokudai.ac.jp
-----

2021年3月22日(月) 21:48 Paolo Giannozzi <p.giannozzi at gmail.com>:

>
> Yes, a fast file system. NFS via Ethernet may still be usable as long as you do not have heavy or frequent I/O. Writing via NFS is especially slow.
>
> For your information: https://gitlab.com/QEF/q-e/-/issues/299
>
> Paolo
>
> On Mon, Mar 22, 2021 at 10:15 AM Takahiro Chiba <takahiro_chiba at eis.hokudai.ac.jp> wrote:
>>
>> Dear Dr. Giannozzi,
>>
>> By the expression "parallel file system", you meant fast file systems
>> such as LUSTRE, not NFS via ethernet, right?
>>
>> ---Sender---
>> Takahiro Chiba
>> Undergraduate at Hokkaido University
>> Expected graduation date: Mar. 2021
>> Expected to be a graduate student at HU from Apr. 2021
>> takahiro_chiba at eis.hokudai.ac.jp
>> -----
>>
>> 2021年3月19日(金) 6:14 Paolo Giannozzi <p.giannozzi at gmail.com>:
>> >
>> > On Thu, Mar 18, 2021 at 6:59 PM Paolo Giannozzi <p.giannozzi at gmail.com> wrote:
>> >
>> >> The "wfcdir"" option is  no longer implemented, if I remember correctly.
>> >
>> >
>> > I didn't remember correctly: it is still implemented. As I mentioned earlier, though, it is expected to work only for a subset of all possible use cases. In particular, it should work for the self-consistent code, but I don't know whether it has ever been tested in NEB.
>> >
>> > Paolo
>> >
>> >
>> >> It was adding unneeded complexity to an already  (too) complex I/O code, it did not work for many use cases, it was useful only for the rather peculiar case of a parallel machine with no parallel file system. Unfortunately the documentation was not updated to reflect its demise.
>> >>
>> >> Paolo
>> >> --
>> >> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
>> >> Univ. Udine, via delle Scienze 206, 33100 Udine, Italy
>> >> Phone +39-0432-558216, fax +39-0432-558222
>> >>
>> >
>> >
>> > --
>> > Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
>> > Univ. Udine, via delle Scienze 206, 33100 Udine, Italy
>> > Phone +39-0432-558216, fax +39-0432-558222
>> >
>> > _______________________________________________
>> > Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
>> > users mailing list users at lists.quantum-espresso.org
>> > https://lists.quantum-espresso.org/mailman/listinfo/users
>> _______________________________________________
>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
>> users mailing list users at lists.quantum-espresso.org
>> https://lists.quantum-espresso.org/mailman/listinfo/users
>
>
>
> --
> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
> Univ. Udine, via delle Scienze 206, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
>
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users