MPI_Bsend-MPI_Recv

In this measurement we use MPI_Bsend for sending and MPI_Recv for receiving. Here we can fix the overhead of managing user-defined buffers.

• Pattern: Point-to-Point varied over message length.
• default values: 8 nodes, max. / act. time for suite disabled/0.34 min.
• Max. allowed standard error is 3.00 %.

	message length
X axis scale	logarithmic
Param. refinement	automatic x wide adaption
Argument range	1 - 4193264 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: 226.2 $\mu$s
• Bandwidth at msg. length 0 units: 0 units/s
• Latency at msg. length 4094.98 * 1024 units: 802592 $\mu$s
• Bandwidth at msg. length 4094.98 * 1024 units: 10204.4 * 1024 units/s