#1528: check that mainstream machines use correct thread binding ------------------------+--------------------------------------------------- Reporter: rhaas | Owner: eschnett Type: task | Status: new Priority: major | Milestone: Component: SimFactory | Version: development version Keywords: | ------------------------+--------------------------------------------------- it is possible that systems that initialize themselves before hwloc (eg MPI) bind to the wrong CPU core unless (the equivalent of) numactl is used.
For an example on bluewaters see #1527.
#1528: check that mainstream machines use correct thread binding -------------------------+-------------------------------------------------- Reporter: rhaas | Owner: eschnett Type: task | Status: new Priority: major | Milestone: Component: SimFactory | Version: development version Resolution: | Keywords: -------------------------+--------------------------------------------------
Comment (by eschnett):
This had a significant performance impact on Blue Waters (from 14 M/hr to 17 M/hr).
#1528: check that mainstream machines use correct thread binding -------------------------+-------------------------------------------------- Reporter: rhaas | Owner: eschnett Type: task | Status: new Priority: major | Milestone: Component: SimFactory | Version: development version Resolution: | Keywords: -------------------------+--------------------------------------------------
Comment (by eschnett):
We need to check the following systems:
- bluewaters (done) - carver - datura - gordon - hopper - kraken - lonestar - orca - pandora - philip - queenbee - shelob - stampede - stampede-mic - titan - trestles - vesta - zwicky
Please edit this text, appending e.g. "(done)" when a machine has been checked.
#1528: check that mainstream machines use correct thread binding -------------------------+-------------------------------------------------- Reporter: rhaas | Owner: eschnett Type: task | Status: new Priority: major | Milestone: Component: SimFactory | Version: development version Resolution: | Keywords: -------------------------+--------------------------------------------------
Comment (by hinder):
Can hwloc be given a parameter to abort with an error message if the thread binding is not correct? If it doesn't know what should be correct, this could be passed in using some environment variables. This would simplify the testing.
#1528: check that mainstream machines use correct thread binding -------------------------+-------------------------------------------------- Reporter: rhaas | Owner: eschnett Type: task | Status: new Priority: major | Milestone: Component: SimFactory | Version: development version Resolution: | Keywords: -------------------------+--------------------------------------------------
Comment (by eschnett):
Yes, such a parameter would be possible. It is in general not clear (without experimentation) which thread bindings are optimal, but we can flag the really bad cases where processes are not bound to the correct socket.
trac@lists.einsteintoolkit.org
-
Einstein Toolkit