APPS/netpipe/dox/netpipe.1

.\" -*- nroff -*-
.\"
.\" NetPIPE -- Network Protocol Independent Performance Evaluator.
.\" Copyright 1997, 1998 Iowa State University Research Foundation, Inc.
.\"
.\" This program is free software; you can redistribute it and/or modify
.\" it under the terms of the GNU General Public License as published by
.\" the Free Software Foundation.  You should have received a copy of the
.\" GNU General Public License along with this program; if not, write to the
.\" Free Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
.\"
.\" netpipe.1
.\" Created: Mon Jun 15 1998 by Guy Helmer
.\" Rewritten:   Jun  1 2004 by Dave Turner
.\"
.\" $Id: netpipe.1,v 1.2 2006-07-19 14:08:57 randall Exp $
.\"
.TH netpipe 1 "June 1, 2004" "NetPIPE" "netpipe"

.SH NAME
NetPIPE \- 
.IB Net work
.IB P rotocol
.IB I ndependent
.IB P erformance
.IB E valuator

.SH SYNOPSIS
.B NPtcp
[\c
.BI \-h \ receiver_hostname\fR\c
]
[\c
.BI \-b \ TCP_buffer_sizes\fR\c
]
[options]

.PP

mpirun
[\c
.BI \-machinefile \ hostlist\fR\c
]
-np 2
.B NPmpi
[-a] [-S] [-z] [options]

.PP

mpirun
[\c
.BI \-machinefile \ hostlist\fR\c
]
-np 2
.B NPmpi2
[-f] [-g] [options]


.PP

.B NPpvm
[options]

See the TESTING sections below for a more complete description of
how to run NetPIPE in each environment.
The OPTIONS section describes the general options available for
all modules.
See the README file from the tar-ball at 
http://www.scl.ameslab.gov/Projects/NetPIPE/ for documentation on
the InfiniBand, GM, SHMEM, LAPI, and memcpy modules.

.SH DESCRIPTION
.PP

.B NetPIPE
uses a simple series of ping-pong tests over a range of message
sizes to provide a complete measure of the performance of a network.
It bounces messages of increasing size between two processes, whether across a 
network or within an SMP system. 
Message sizes are chosen at regular intervals, and with slight perturbations, 
to provide a complete evaluation of the communication system. 
Each data point involves many ping-pong tests to provide an accurate timing. 
Latencies are calculated by dividing the round trip time in half for small 
messages ( less than 64 Bytes ). 
.PP
The communication time for small messages is dominated by the 
overhead in the communication layers, meaning that the transmission
is latency bound.
For larger messages, the communication rate becomes bandwidth limited by 
some component in
the communication subsystem (PCI bus, network card link, network switch).
.PP
These measurements can be done at the message-passing layer 
(MPI, MPI-2, and PVM) or at the native communications layers
that that run upon (TCP/IP, GM for Myrinet cards, InfiniBand,
SHMEM for the Cray T3E systems, and LAPI for IBM SP systems).
Recent work is being aimed at measuring some internal system properties
such as the memcpy module that measures the internal memory copy rates,
or a disk module under development that measures the performance
to various I/O devices.
.PP

Some uses for NetPIPE include:
.RS
.PP
Comparing the latency and maximum throughput of various network cards.
.PP
Comparing the performance between different types of networks.
.PP
Looking for inefficiencies in the message-passing layer by comparing it
to the native communication layer.
.PP
Optimizing the message-passing layer and tune OS and driver parameters
for optimal performance of the communication subsystem.

.RE
.PP

.B NetPIPE
is provided with many modules allowing it to interface with a wide
variety of communication layers.
It is fairly easy to write new interfaces for other reliable protocols
by using the existing modules as examples.



.SH TESTING TCP
.PP
NPtcp can now be launched in two ways, by manually starting NPtcp on
both systems or by using a nplaunch script.  To manually start NPtcp,
the NetPIPE receiver must be 
started first on the remote system using the command:
.PP
.Ex
NPtcp [options]
.Ee
.PP
then the primary transmitter is started on the local system with the
command
.PP
.Ex
NPtcp \-h 
.I receiver_hostname
[options]
.Ee
.PP
Any options used must be the same on both sides.  The \-P parameter
can be used to override the default port number.  This is helpful when
running several streams through a router to a single endpoint.

The nplaunch script uses ssh to launch the remote receiver
before starting the local transmitter.  To use rsh, simply change
the nplaunch script.
.PP
.Ex
nplaunch NPtcp -h 
.I receiver_hostname
[options]
.Ee
.PP
The
.BI \-b \ TCP_buffer_sizes\fR\c
 option sets the TCP socket buffer size, which can greatly influence
the maximum throughput on some systems.  A throughput graph that
flattens out suddenly may be a sign of the performance being limited
by the socket buffer sizes.
.PP
Several other protocols are testable in the same way as TCP.  These
include TCP6 (TCP over IPv6), SCTP and IPX.  They are started in the
same way but the program names are NPtcp6, NPsctp, and NPipx
respectively.


.SH TESTING MPI and MPI-2
.PP
Use of the MPI interface for NetPIPE depends on the MPI implementation
being used.  
All will require the number of processes to be specified, usually
with a 
.I -np 2 
argument.  Clusters environments may require a list of the 
hosts being used, either during initialization of MPI (during lamboot
for LAM-MPI) or when each job is run (using a -machinefile argument
for MPICH).
For LAM-MPI, for example, put the list of hosts in hostlist then boot LAM 
and run NetPIPE using:
.PP
.Ex
lamboot -v -b 
.I hostlist
.PP
mpirun \-np 2 NPmpi [NetPIPE options]
.Ee
.PP

For MPICH use a command like:
.PP
.Ex
mpirun \-machinefile 
.I hostlist 
\-np 2 NPmpi [NetPIPE options]
.Ee
.PP

To test the 1-sided communications of the MPI-2 standard, compile
using:
.PP
.Ex
.B make mpi2
.Ee
.PP
Running as described above and MPI will use 1-sided MPI_Put()
calls in both directions, with each receiver blocking until the
last byte has been overwritten before bouncing the message back.
Use the 
.I -f
option to force usage of a fence to block rather than an overwrite
of the last byte.
The 
.I -g
option will use MP_Get() functions to transfer the data rather than
MP_Put().


.SH TESTING PVM
.PP
Start the pvm system using:
.PP
.Ex
pvm
.Ee
.PP
and adding a second machine with the PVM command
.PP
.Ex
add 
.I receiver_hostname
.Ee
.PP
Exit the PVM command line interface using quit, then run the PVM NetPIPE
receiver on one system with the command:
.PP
.Ex
NPpvm [options]
.Ee
.PP
and run the TCP NetPIPE transmitter on the other system with the
command:
.PP
.Ex
NPpvm -h 
.I receiver hostname
[options]
.Ee
.PP
Any options used must be the same on both sides.
The nplaunch script may also be used with NPpvm as described above
for NPtcp.

.SH TESTING METHODOLOGY
.PP
.B NetPIPE
tests network performance by sending a number of messages at each
block size, starting from the lower bound on the message sizes.

The message size is incremented until the upper bound on the message size is
reached or the time to transmit a block exceeds one second, which ever
occurs first.  Message sizes are chosen at regular intervals, and for
slight perturbations from them to provide a more complete evaluation
of the communication subsystem.
.PP
The 
.B NetPIPE\c
 output file may be graphed using a program such as
.B gnuplot(1)\.
The output file contains three columns: the number of bytes in the block,
the transfer rate in bits per second, and
the time to transfer the block (half the round-trip time).
The first two columns are normally used to graph the throughput
vs block size, while the third column provides the latency.
For example, the 
.B throughput versus block size
graph can be created by graphing bytes versus bits per second.
Sample
.B gnuplot(1)
commands for such a graph would be
.PP
.Ex
set logscale x
.Ee
.PP
.Ex
plot "np.out"
.Ee

.ne 5
.SH OPTIONS
.TP
.B \-a
asynchronous mode: prepost receives (MPI, IB modules)
.ne 3
.TP
.BI \-b \ \fITCP_buffer_sizes\fR
Set the send and receive TCP buffer sizes (TCP module only).
.ne 3

.TP
.B \-B
Burst mode where all receives are preposted at once (MPI, IB modules).
.ne 3

.TP
.B \-f
Use a fence to block for completion (MPI2 module only).
.ne 3

.TP
.B \-g
Use MPI_Get() instead of MPI_Put() (MPI2 module only).
.ne 3

.TP
.BI \-h \ \fIhostname\fR
Specify the name of the receiver host to connect to (TCP, PVM, IB, GM).
.ne 3

.TP
.B \-I
Invalidate cache to measure performance without cache effects (mostly affects 
IB and memcpy modules).
.ne 3

.TP
.B \-i
Do an integrity check instead of a performance evaluation.
.ne 3

.TP
.BI \-l \ \fIstarting_msg_size\fR
Specify the lower bound for the size of messages to be tested.
.ne 3
.TP

.TP
.BI \-n \ \fInrepeats\fR
Set the number of repeats for each test to a constant.
Otherwise, the number of repeats is chosen to provide an accurate
timing for each test.  Be very careful if specifying a low number
so that the time for the ping-pong test exceeds the timer accuracy.
.ne 3

.TP
.BI \-O \ \fIsource_offset,dest_offset\fR
Specify the source and destination offsets of the buffers from perfect 
page alignment.
.ne 3
.TP

.BI \-o \ \fIoutput_filename\fR
Specify the output filename (default is np.out).
.ne 3

.TP
.BI \-p \ \fIperturbation_size\fR
NetPIPE chooses the message sizes at regular intervals, increasing them
exponentially from the lower boundary to the upper boundary.
At each point, it also tests perturbations of 3 bytes above and 3 bytes
below each test point to find idiosyncrasies in the system.
This perturbation value can be changed using the 
.I -p
option, or turned
off using
.I -p 
.I 0
.B .
.ne 3

.TP
.B \-r
This option resets the TCP sockets after every test (TCP module only).
It is necessary for some streaming tests to get good measurements
since the socket window size may otherwise collapse.
.ne 3

.TP
.B \-s
Set streaming mode where data is only transmitted in one direction. 
.ne 3

.TP
.B \-S
Use synchronous sends (MPI module only).
.ne 3

.TP
.BI \-u \ \fIupper_bound\fR
Specify the upper boundary to the size of message being tested.  
By default, NetPIPE will stop when
the time to transmit a block exceeds one second. 

.TP
.B \-z
Receive messages using MPI_ANY_SOURCE (MPI module only)
.ne 3

.TP
.B \-2
Set bi-directional mode where both sides send and receive at the
same time (supported by most modules).
You may need to use 
.I -a
to choose asynchronous communications for MPI to avoid freeze-ups.
For TCP, the maximum test size will be limited by the TCP
buffer sizes.
.ne 3

.ne 3
.SH FILES
.TP
.I np.out
Default output file for
.BR NetPIPE .
Overridden by the
.B \-o
option.

.SH AUTHOR
.PP
The original NetPIPE core plus TCP and MPI modules were written by 
Quinn Snell, Armin Mikler, Guy Helmer, and John Gustafson.
NetPIPE is currently being developed and maintained by Dave Turner
with contributions from many students (Bogdan Vasiliu, Adam Oline,
Xuehua Chen, and Brian Smith).  

.PP
Send comments/bug-reports to:
.I
<netpipe@scl.ameslab.gov>.
.PP
Additional information about
.B NetPIPE
can be found on the World Wide Web at
.I http://www.scl.ameslab.gov/Projects/NetPIPE/

.SH BUGS
As of version 3.6.1, there is a bug that causes NetPIPE to segfault on 
RedHat Enterprise systems. I will debug this as soon as I get access to a 
few such systems.  -Dave Turner (turner@ameslab.gov)