<div>Hi Wei,</div><div><br></div><div>there are a number of methods in CP2K that can take advantage of GPUs. For more details see: https://www.cp2k.org/gpu</div><div><br></div><div>However, there are also a number of know issues which we're still working on:<br></div><div> - The grid code does not perform well for large basis sets: https://github.com/cp2k/cp2k/issues/1785</div><div> - DBCSR struggles with multi-GPU.</div><div><br></div><div>-Ole</div><div><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">On Wednesday, June 22, 2022 at 5:14:53 PM UTC+2 yushu...@gmail.com wrote:<br/></div><blockquote class="gmail_quote" style="margin: 0 0 0 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">Hi,<div>I just found the NGC container for CP2k was released and tried to use it for a MD simulation at Nvidia DXG A100 system but did not find much speedup as expected. Is there any benchmark of the speedup by using CPUs compared to GPUs?</div><div>Thanks.</div><div><br></div><div>Wei</div></blockquote></div>
<p></p>
-- <br />
You received this message because you are subscribed to the Google Groups "cp2k" group.<br />
To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:cp2k+unsubscribe@googlegroups.com">cp2k+unsubscribe@googlegroups.com</a>.<br />
To view this discussion on the web visit <a href="https://groups.google.com/d/msgid/cp2k/f8f29eaf-fa22-409f-94ce-7c2be92c1990n%40googlegroups.com?utm_medium=email&utm_source=footer">https://groups.google.com/d/msgid/cp2k/f8f29eaf-fa22-409f-94ce-7c2be92c1990n%40googlegroups.com</a>.<br />