MPI_Isend-length

This measurement is similar to the MPI_Send-length measurement. It only differs in using MPI_Isend instead of MPI_Ssend. The non-blocking sending will be faster than the blocking variants, if the network allowes.

• Pattern: Master-Worker varied over message length.
• default values: 8 nodes, chunks 10, max. / act. time for suite disabled/0.25 min.
• Max. allowed standard error is 3.00 %.

	message length
X axis scale	logarithmic
Param. refinement	automatic x wide adaption
Argument range	1 - 4194304 units (s. below)
Argument stepwidth	1.414214

• Units: bytes in type: basetype 1 (MPI_INT), send-datatype 20: size 4 Bytes, extent 4 Bytes (user defined basetype 1 (MPI_INT)), receive-datatype 20: size 4 Bytes, extent 4 Bytes (user defined basetype 1 (MPI_INT))

• Latency at msg. length 0 units: 108.1 $\mu$s
• Bandwidth at msg. length 0 units: 0 units/s
• Latency at msg. length 4096 * 1024 units: 112998 $\mu$s
• Bandwidth at msg. length 4096 * 1024 units: 362484 * 1024 units/s