Ya -- sorry about that. I was writing the libNBC performance analyzer at the time and had the parser wrong. More info coming about libNBC shortly -- it outputs a whole pile of data that we're currently not tracking. We'll need some DB/reporter/client-side work to track it all (analogous to, but easier than what we were previously talking about for SKaMPI).

On Oct 7, 2007, at 7:33 PM, Josh Hursey wrote:

FYI:
I just got this error from MTT submit.php. It seems that the
message_size array is a bit corrupted.

On Oct 7, 2007, at 11:25 AM, jjhur...@open-mpi.org wrote:


SQL QUERY: INSERT INTO latency_bandwidth
        (latency_bandwidth_id, message_size, latency_min, latency_avg,
latency_max, bandwidth_min, bandwidth_avg, bandwidth_max) VALUES
        ('138988', '{on,with,NBC_Test,each,
1,2,4,8,16,32,64,128,256,512,1024,2048,4096,8192,16384,32768,65536,13 1
072,262144,524288,1048576}', DEFAULT, '{,,,the,
9.06,9.06,10.01,10.01,9.06,10.01,9.06,10.01,10.01,10.97,11.92,11.92,2 3
.13,29.09,43.15,67.95,120.88,216.96,471.83,1347.06,4244.09}',
DEFAULT, DEFAULT, DEFAULT, DEFAULT)
SQL ERROR: ERROR:  invalid input syntax for integer: "on"
SQL ERROR:

_______________________________________________
mtt-devel mailing list
mtt-de...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/mtt-devel


--
Jeff Squyres
Cisco Systems

Reply via email to