<div dir="ltr">Dear Prof. Blaha,<div><br></div><div style> Thank you very much for the explanations. They will be very useful !!</div><div style> All the best,</div><div style> Luis</div></div><div class="gmail_extra">
<br><br><div class="gmail_quote">2013/9/12 Peter Blaha <span dir="ltr"><<a href="mailto:pblaha@theochem.tuwien.ac.at" target="_blank">pblaha@theochem.tuwien.ac.at</a>></span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
This depends a lot on what you want to do and how much money you have.<br>
<br>
The single-core speed of a fast I7 is at least as good (or faster) than most Xeons, and they are MUCH cheaper. So for all systems up to 64-100 atoms/cell, where you need several k-points, a small cluster of I7 cpus<br>
/4 cores, or more expensive 6 core) (with GB-network and a common NFS file system) is for sure the fastest platform and in particular has by FAR the best price/performance ratio (One powerful I7 computer may cost about 1000 Euro). For bigger clusters, a drawback can be the large "space" to put all PCs on a big shelf ....), but if you have less than 10000 Euros, this is probably the best choice.<br>
<br>
However, Xeons can be coupled (2-4 Xeons) to a "single multicore computer (eg. 16 cores)", which may work with mpi and can be used to handle systems up to 200-300 atoms. They also can be bought in small boxes and may fit in a single 19 inch cabinet. But of course such systems are much more expensive. From what I said above it should be clear, that it is completely useless to buy a "single 4 core Xeon computer".<br>
<br>
The next step would be to buy an Infiniband switch+cards and couple your PCs with this fast network to a powerful multinode mpi-cluster. Since the switch/cards are fairly expensive, on usually takes here Xeons as platform. However, you need to know how to install/configure the software properly. I've seen such clusters even in computing centers, which were completely useless, because the network/mpi was instable and jobs would crash randomly every couple of hours .....<br>
<br>
Our strategy:<br>
i) We have a GB-networked cluster with Intel I7 computers (which we maintain our-self and this cluster includes also all the user- workstations) and do all the calculations for systems up to 64 atoms/cell on these systems.<br>
2) For bigger systems we go to our University computer-center and run there with a PBS queuing system. This has the advantage that we do not need to care about the installation of the infiniband network nor the mpi-infrastructure (but we use always intel-mpi together with ifort/mkl).<div class="im">
<br>
<br>
On 09/11/2013 06:31 PM, Luis Ogando wrote:<br>
</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">
Dear Prof. Blaha,<br>
<br>
Just for curiosity, what processor did you buy ?<br>
Is the Xeon family better than the i7 one for WIEN2k calculations ?<br>
All the best,<br>
Luis<br>
<br>
<br>
2013/9/11 Peter Blaha <<a href="mailto:pblaha@theochem.tuwien.ac.at" target="_blank">pblaha@theochem.tuwien.ac.at</a><br></div>
<mailto:<a href="mailto:pblaha@theochem.tuwien.ac.at" target="_blank">pblaha@theochem.<u></u>tuwien.ac.at</a>>><div class="im"><br>
<br>
I don't know what "latest" means. We use the latest one installed on<br>
our supercomputers (4.1.1.036)<br>
<br>
I have not seen any significant change with mpi in the last years.<br>
<br>
PS: I just got info that we have now a new ifort available for<br>
download ...<br>
<br>
<br>
On 09/11/2013 05:00 PM, Laurence Marks wrote:<br>
<br>
Thanks.<br>
<br>
One thing I will add/ask concerning the parallelization, the latest<br>
impi seems to be substantially better -- have you tried it? I<br>
have not<br>
just noticed this with Wien2k, but I am told that others have seen<br>
improvements in other codes.<br>
<br>
On Wed, Sep 11, 2013 at 9:42 AM, Peter Blaha<br>
<<a href="mailto:pblaha@theochem.tuwien.ac.at" target="_blank">pblaha@theochem.tuwien.ac.at</a><br></div><div><div class="h5">
<mailto:<a href="mailto:pblaha@theochem.tuwien.ac.at" target="_blank">pblaha@theochem.<u></u>tuwien.ac.at</a>>> wrote:<br>
<br>
Before buying a couple of new computers, I was asking myself<br>
the same<br>
question and discussed this with some people of our<br>
computing departments.<br>
<br>
The conclusions:<br>
a) potentially very good, but in practice very questionable,<br>
because for<br>
most application you cannot get out the real speed (10 times<br>
faster than<br>
an Intel I7). This is true, even for many lapack/mkl<br>
subroutines where<br>
it "should" work better.<br>
They told me to "wait", until the mkl becomes better<br>
(hopefully). I'm<br>
not too optimistic, when you see how badly the<br>
mkl-parallelization of<br>
multicore machines is working (2 cores is very good, but 4<br>
or more is<br>
already very bad).<br>
<br>
b) The nature of our problem (big eigenvalue problem): A "fast<br>
processor" is useful only for large problems --> large memory.<br>
You can buy Phi coprocessors now with quite some large<br>
memory, but then<br>
they are terrible expensive (and 5 "normal" PCs are faster<br>
and cheaper)<br>
<br>
c) the hardware design has a VERY slow communication between<br>
main-memory<br>
and Phi-memory. This makes also parallelization over several<br>
PHI-nodes<br>
via mpi not really possible (if you need any significant<br>
data transfer,<br>
like for an eigenvalue problem).<br>
<br>
Thus I did not buy it.<br>
<br>
However, if anybody has access and time to try out WIEN2k on<br>
PHis, I'd<br>
would be very interested in getting feedback. (Maybe these<br>
computer-people were not good enough ....)<br>
<br>
PS: I know from G.Kresse that they had some time ago (maybe<br>
2 years ?)<br>
an expert from NVIDIA with them. After 2 weeks of porting<br>
VASP to these<br>
GPUs by this expert, VASP on the GPU was "almost as fast" as<br>
on an Intel<br>
I7 processor.<br>
<br>
<br>
On 09/11/2013 04:16 PM, Laurence Marks wrote:<br>
<br>
Anyone know if these will be viable with Wien2k (mpi,<br>
i.e. large problems)?<br>
<br>
<br>
--<br>
<br>
P.Blaha<br></div></div>
------------------------------<u></u>__----------------------------<u></u>--__--------------<div class="im"><br>
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060<br>
Vienna<br></div>
Phone: <a href="tel:%2B43-1-58801-165300" value="+43158801165300" target="_blank">+43-1-58801-165300</a> <tel:%2B43-1-58801-165300><br>
FAX: <a href="tel:%2B43-1-58801-165982" value="+43158801165982" target="_blank">+43-1-58801-165982</a> <tel:%2B43-1-58801-165982><br>
Email: <a href="mailto:blaha@theochem.tuwien.ac.at" target="_blank">blaha@theochem.tuwien.ac.at</a><br>
<mailto:<a href="mailto:blaha@theochem.tuwien.ac.at" target="_blank">blaha@theochem.tuwien.<u></u>ac.at</a>> WWW:<br>
<a href="http://info.tuwien.ac.at/__theochem/" target="_blank">http://info.tuwien.ac.at/__<u></u>theochem/</a><br>
<<a href="http://info.tuwien.ac.at/theochem/" target="_blank">http://info.tuwien.ac.at/<u></u>theochem/</a>><br>
------------------------------<u></u>__----------------------------<u></u>--__--------------<br>
______________________________<u></u>___________________<br>
Wien mailing list<br>
Wien@zeus.theochem.tuwien.ac._<u></u>_at<br>
<mailto:<a href="mailto:Wien@zeus.theochem.tuwien.ac.at" target="_blank">Wien@zeus.theochem.<u></u>tuwien.ac.at</a>><br>
<a href="http://zeus.theochem.tuwien." target="_blank">http://zeus.theochem.tuwien.</a>__<a href="http://ac.at/mailman/listinfo/wien" target="_blank"><u></u>ac.at/mailman/listinfo/wien</a><div class="im"><br>
<<a href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien" target="_blank">http://zeus.theochem.tuwien.<u></u>ac.at/mailman/listinfo/wien</a>><br>
SEARCH the MAILING-LIST at:<br></div>
<a href="http://www.mail-archive.com/__wien@zeus.theochem.tuwien.ac.__at/index.html" target="_blank">http://www.mail-archive.com/__<u></u>wien@zeus.theochem.tuwien.ac._<u></u>_at/index.html</a><br>
<<a href="http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html" target="_blank">http://www.mail-archive.com/<u></u>wien@zeus.theochem.tuwien.ac.<u></u>at/index.html</a>><br>
<br>
<br>
<br>
<br>
<br>
--<br>
<br>
P.Blaha<br>
------------------------------<u></u>__----------------------------<u></u>--__--------------<div class="im"><br>
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna<br></div>
Phone: <a href="tel:%2B43-1-58801-165300" value="+43158801165300" target="_blank">+43-1-58801-165300</a> <tel:%2B43-1-58801-165300><br>
FAX: <a href="tel:%2B43-1-58801-165982" value="+43158801165982" target="_blank">+43-1-58801-165982</a> <tel:%2B43-1-58801-165982><br>
Email: <a href="mailto:blaha@theochem.tuwien.ac.at" target="_blank">blaha@theochem.tuwien.ac.at</a><br>
<mailto:<a href="mailto:blaha@theochem.tuwien.ac.at" target="_blank">blaha@theochem.tuwien.<u></u>ac.at</a>> WWW:<br>
<a href="http://info.tuwien.ac.at/__theochem/" target="_blank">http://info.tuwien.ac.at/__<u></u>theochem/</a><br>
<<a href="http://info.tuwien.ac.at/theochem/" target="_blank">http://info.tuwien.ac.at/<u></u>theochem/</a>><br>
------------------------------<u></u>__----------------------------<u></u>--__--------------<br>
______________________________<u></u>___________________<br>
Wien mailing list<br>
Wien@zeus.theochem.tuwien.ac._<u></u>_at<br>
<mailto:<a href="mailto:Wien@zeus.theochem.tuwien.ac.at" target="_blank">Wien@zeus.theochem.<u></u>tuwien.ac.at</a>><br>
<a href="http://zeus.theochem.tuwien." target="_blank">http://zeus.theochem.tuwien.</a>__<a href="http://ac.at/mailman/listinfo/wien" target="_blank"><u></u>ac.at/mailman/listinfo/wien</a><div class="im"><br>
<<a href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien" target="_blank">http://zeus.theochem.tuwien.<u></u>ac.at/mailman/listinfo/wien</a>><br>
SEARCH the MAILING-LIST at:<br></div>
<a href="http://www.mail-archive.com/__wien@zeus.theochem.tuwien.ac.__at/index.html" target="_blank">http://www.mail-archive.com/__<u></u>wien@zeus.theochem.tuwien.ac._<u></u>_at/index.html</a><br>
<<a href="http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html" target="_blank">http://www.mail-archive.com/<u></u>wien@zeus.theochem.tuwien.ac.<u></u>at/index.html</a>><div class="im"><br>
<br>
<br>
<br>
<br>
______________________________<u></u>_________________<br>
Wien mailing list<br>
<a href="mailto:Wien@zeus.theochem.tuwien.ac.at" target="_blank">Wien@zeus.theochem.tuwien.ac.<u></u>at</a><br>
<a href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien" target="_blank">http://zeus.theochem.tuwien.<u></u>ac.at/mailman/listinfo/wien</a><br>
SEARCH the MAILING-LIST at: <a href="http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html" target="_blank">http://www.mail-archive.com/<u></u>wien@zeus.theochem.tuwien.ac.<u></u>at/index.html</a><br>
<br>
</div></blockquote><div class="HOEnZb"><div class="h5">
<br>
-- <br>
<br>
P.Blaha<br>
------------------------------<u></u>------------------------------<u></u>--------------<br>
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna<br>
Phone: <a href="tel:%2B43-1-58801-165300" value="+43158801165300" target="_blank">+43-1-58801-165300</a> FAX: <a href="tel:%2B43-1-58801-165982" value="+43158801165982" target="_blank">+43-1-58801-165982</a><br>
Email: <a href="mailto:blaha@theochem.tuwien.ac.at" target="_blank">blaha@theochem.tuwien.ac.at</a> WWW: <a href="http://info.tuwien.ac.at/theochem/" target="_blank">http://info.tuwien.ac.at/<u></u>theochem/</a><br>
------------------------------<u></u>------------------------------<u></u>--------------<br>
______________________________<u></u>_________________<br>
Wien mailing list<br>
<a href="mailto:Wien@zeus.theochem.tuwien.ac.at" target="_blank">Wien@zeus.theochem.tuwien.ac.<u></u>at</a><br>
<a href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien" target="_blank">http://zeus.theochem.tuwien.<u></u>ac.at/mailman/listinfo/wien</a><br>
SEARCH the MAILING-LIST at: <a href="http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html" target="_blank">http://www.mail-archive.com/<u></u>wien@zeus.theochem.tuwien.ac.<u></u>at/index.html</a><br>
</div></div></blockquote></div><br></div>