[HPGMG Forum] HPGMG release v0.1

Jed Brown jed at jedbrown.org
Sat Jun 7 22:39:02 UTC 2014


Barry Smith <bsmith at mcs.anl.gov> writes:

>    The use of multicore processor == sockets as the independent
>    variable in the plot of aggregate performance seems
>    arbitrary. Though people should not use this kind of plot to
>    compare machines they will. Now just change sockets to nodes and
>    boom suddenly the machines compare very differently (since some
>    systems have two sockets per node and some one). 

This only changes the x axis, which is a fairly arbitrary measure of how
much machine is being used.  The y axis is a figure of merit (probably
not The Metric).

>    Should cores be used instead? Or hardware threads? Or cores scaled
>    by their clock speed? Or hardware floating point units (scaled by
>    clock speed?) ?  Or number of instruction decorders? 

These all have the same problem.

>    Power usage?  

I think power usage is a more useful normalization of machine size, but
number of nodes or number of sockets is the most common way that users
of a particular machine think about a partition size.

>    Cost? etc etc. Maybe have a dynamic plot where one can switch the
>    independent variable by selecting from a menu or moving the mouse
>    over choices ….?

That would be cool, but my D3.js skills will have to improve to make
that interactive plot.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 818 bytes
Desc: not available
URL: <https://hpgmg.org/lists/archives/hpgmg-forum/attachments/20140608/d1879a71/attachment-0001.bin>


More information about the HPGMG-Forum mailing list