<html>
<body>
At 09:38 AM 2/23/2008, Richard Graham wrote:<br>
<blockquote type=cite class=cite cite=""><font face="Verdana">So I think
we are some what talking past each other. I think that what you
really care about<br>
with respect to communications errors is information on messages
that have not completed,<br>
and, more important, can’t complete ? Is this correct ?<br>
I have focused more on errors that occur, but the low-level can handle
them and does not need<br>
to pass information about them back up to the user. I believe
this is where we said that we<br>
may want to be able to give the app some indication on performance
degradation, at their<br>
request. Is this correct ?<br>
</font></blockquote><br>
Exactly. My point was that the former belongs in the transactional memory
API, while the latter belongs in the QoS API. Kannan and I are tasked
with drafting something for the latter. One thing to note though is that
identical low-level events may fall under either API. In particular, a
single-bitflip may result in a message drop in one MPI implementation and
a seamless recovery with minor performance degradation in another. The
MPI implementation gets to choose the mapping between low-level events
and their high-level manifestations. Right now we are just trying to
define a reasonable interface for the high-level manifestations.<br>
<x-sigsep><p></x-sigsep>
Greg Bronevetsky<br>
Post-Doctoral Researcher<br>
1028 Building 451<br>
Lawrence Livermore National Lab<br>
(925) 424-5756<br>
bronevetsky1@llnl.gov</body>
</html>