<div dir="ltr"><div id="m_-4887640929092430203DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2">Thank for your response,</div><div id="m_-4887640929092430203DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2"><br></div><div id="m_-4887640929092430203DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2">I ran it again with 6.5 (couldn't install 6.6a1), it uses the serial eigensolver.</div><div id="m_-4887640929092430203DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2"><br></div><div id="m_-4887640929092430203DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2">now I get :</div><div id="m_-4887640929092430203DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2"> Band Structure Calculation<br> Davidson diagonalization with overlap<br><br> Computing kpt #: 1 of 9 on this pool<br> Really copied g2kin H->D<br> Really copied evc H->D<br> Really copied et H->D<br><br> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br> Error in routine cegterg (1):<br> cannot allocate vc_d<br> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br><br> stopping ...<br></div><div><br></div>I attach input and output<div><br></div><div>I'll put the rest on gitlab<br><div><br></div><div>Thank you, </div><div>Sara<br><div dir="ltr"><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">Il giorno dom 30 ago 2020 alle ore 23:18 Pietro Bonfa <<a href="mailto:pietro.bonfa@unipr.it" target="_blank">pietro.bonfa@unipr.it</a>> ha scritto:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Dear Sara,<br>
<br>
I'd suggest checking the following:<br>
<br>
1. verify that the serial eigensolver is used (it's written at the<br>
beginning of the output);<br>
<br>
2. use the latest version (6.6a1) that will correctly report problems<br>
with memory allocations during the iterative diagonalization.<br>
<br>
Could you please also open an issue at<br>
<a href="https://gitlab.com/QEF/q-e-gpu/-/issues" rel="noreferrer" target="_blank">https://gitlab.com/QEF/q-e-gpu/-/issues</a> and attach the input, the<br>
pseudopotentials and the job script that you are using?<br>
<br>
Thank you,<br>
kind regards,<br>
Pietro<br>
<br>
<br>
<br>
On 8/29/20 6:33 PM, Sara Postorino wrote:<br>
> Hi QE users,<br>
><br>
> I am running PW on Marconi100 and experiencing problems during<br>
> digonalization. I am using version 6.5 (autoload of the modules on m100).<br>
> My system is a MoTe2 bilayer k mesh 39x39x1 with many bands due to the<br>
> fact that I will do a GW calculation on top of it. (The calculation<br>
> works if I do not add many bands)<br>
> I tried with 4000 and 3000 bands using Davidson diagonalization running<br>
> on 18 nodes:<br>
> Parallel version (MPI & OpenMP), running on 2304 processor cores<br>
> Number of MPI processes: 72<br>
> Threads/MPI process: 32<br>
> When doin the calculation of the first point I get:<br>
><br>
> Really copied g2kin H->D<br>
> Really copied evc H->D<br>
> Really copied et H->D<br>
> Really copied vrs H->D<br>
> dp_memcpy_d2h_c2dinvalid pitch argument 12<br>
><br>
> I also tried with Conjugate gradient algorithm but it gets stuck at<br>
><br>
> Really copied evc H->D<br>
> Really copied et H->D<br>
> Really copied h_diag H->D<br>
> Really copied becp%nc H->D<br>
> Really copied g2kin H->D<br>
> Really copied vrs H->D<br>
><br>
> And here it takes forever. I left it running for more than 1 hour and it<br>
> didn't finish on k point and since I have 147 kpoints the computation<br>
> would be very expensive even if it worked.<br>
><br>
> I also tried to go down to 1000 bands (I need way more) and got<br>
> Really copied g2kin H->D<br>
> Really copied evc H->D<br>
> Really copied et H->D<br>
> Really copied vrs H->D<br>
> zhegvdx_gpu error: cusolverDnZpotrf failed!<br>
><br>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br>
> Error in routine cdiaghg_gpu (1):<br>
> zhegvdx_gpu failed<br>
> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br>
><br>
> Do you have any suggestion on how to fix this issue?<br>
> Thanks<br>
><br>
> Sara Postorino<br>
> PhD student<br>
> University of Rome Tor Vergata<br>
><br>
><br>
> <<a href="https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail" rel="noreferrer" target="_blank">https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail</a>><br>
> Mail priva di virus. <a href="http://www.avast.com" rel="noreferrer" target="_blank">www.avast.com</a><br>
> <<a href="https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail" rel="noreferrer" target="_blank">https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail</a>><br>
><br>
><br>
> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2><br>
><br>
> _______________________________________________<br>
> Quantum ESPRESSO is supported by MaX (<a href="http://www.max-centre.eu/quantum-espresso" rel="noreferrer" target="_blank">http://www.max-centre.eu/quantum-espresso</a><br>
> users mailing list <a href="mailto:users@lists.quantum-espresso.org" target="_blank">users@lists.quantum-espresso.org</a><br>
> <a href="https://lists.quantum-espresso.org/mailman/listinfo/users" rel="noreferrer" target="_blank">https://lists.quantum-espresso.org/mailman/listinfo/users</a><br>
><br>
<br>
Firma il tuo 5 per mille all’Università di Parma e aiuta così i nostri studenti che vogliono realizzare un’esperienza di studio all’estero - Indica 00308780345 nella tua denuncia dei redditi.<br>
_______________________________________________<br>
Quantum ESPRESSO is supported by MaX (<a href="http://www.max-centre.eu/quantum-espresso" rel="noreferrer" target="_blank">www.max-centre.eu/quantum-espresso</a>)<br>
users mailing list <a href="mailto:users@lists.quantum-espresso.org" target="_blank">users@lists.quantum-espresso.org</a><br>
<a href="https://lists.quantum-espresso.org/mailman/listinfo/users" rel="noreferrer" target="_blank">https://lists.quantum-espresso.org/mailman/listinfo/users</a></blockquote></div>
<div id="m_-4887640929092430203DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2"><br> <table style="border-top:1px solid #d3d4de">
<tbody><tr>
<td style="width:55px;padding-top:18px"><a href="https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail" target="_blank"><img src="https://ipmcdn.avast.com/images/icons/icon-envelope-tick-round-orange-animated-no-repeat-v1.gif" alt="" width="46" height="29" style="width:46px;height:29px"></a></td>
<td style="width:470px;padding-top:17px;color:#41424e;font-size:13px;font-family:Arial,Helvetica,sans-serif;line-height:18px">Mail priva di virus. <a href="https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail" style="color:#4453ea" target="_blank">www.avast.com</a> </td>
</tr>
</tbody></table>
<a href="#m_-4887640929092430203_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2" width="1" height="1"></a></div></div></div></div>