Wireless Hacks. 1917 IndustrialStrength Tips and Tools [Electronic resources] نسخه متنی

اینجــــا یک کتابخانه دیجیتالی است

با بیش از 100000 منبع الکترونیکی رایگان به زبان فارسی ، عربی و انگلیسی

Wireless Hacks. 1917 IndustrialStrength Tips and Tools [Electronic resources] - نسخه متنی

Rob Flickenger

| نمايش فراداده ، افزودن یک نقد و بررسی
افزودن به کتابخانه شخصی
ارسال به دوستان
جستجو در متن کتاب
بیشتر
تنظیمات قلم

فونت

اندازه قلم

+ - پیش فرض

حالت نمایش

روز نیمروز شب
جستجو در لغت نامه
بیشتر
لیست موضوعات
توضیحات
افزودن یادداشت جدید










Hack 36 Estimating Network Performance


Just how quickly can you squeeze data through
your AP?


Many people use
online tools like
DSL Reports' Speed test
(http://speedtest.dslreports.com/) to estimate
the performance of their Internet connection.
When run from a machine directly connected to the Internet, this can
give you a fairly good indication of your upload and download
capacity.

This tool becomes less useful when trying to estimate the available
bandwidth on other networks. For example, on a large wireless
network, it is useful to measure the actual capacity of a network
link regardless of the speed of the Internet connection. One useful
utility for measuring performance is
iperf. It is a simple, freely available tool
that will run on Linux, BSD, OS X, and even Windows. You can download
it online at http://dast.nlanr.net/Projects/Iperf/.

In order to measure performance, it needs to be used in pairs (one
instance at either end of a link). On one end of the link to be
measured, start up iperf in server mode:

rob@livia:~$ iperf -s

Note that it doesn't matter which end is used as the
"server," as both upload and
download speeds will be tested. On the other end of the link, run
iperf in client mode, specifying the server to
be tested:

rob@caligula:~$ iperf -c livia -r
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 32.0 KByte (default)
------------------------------------------------------------
------------------------------------------------------------
Client connecting to livia, TCP port 5001
TCP window size: 32.5 KByte (default)
------------------------------------------------------------
[ 4] local 10.15.6.33 port 50421 connected with 10.15.6.4 port 5001
[ ID] Interval Transfer Bandwidth
[ 4] 0.0-10.2 sec 2.95 MBytes 2.43 Mbits/sec
[ 4] local 10.15.6.33 port 5001 connected with 10.15.6.4 port 60977
[ ID] Interval Transfer Bandwidth
[ 4] 0.0-10.0 sec 3.09 MBytes 2.60 Mbits/sec
rob@caligula:~$

By default, iperf uses port 5001 for its
communications. If this port is in use, you can specify a different
one with the -p switch on both sides:

rob@livia:~$ iperf -s -p 30000

You can specify a different port and on the client as well:

rob@caligula:~$ iperf -c livia -r -p 30000
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 32.0 KByte (default)
...

If you don't want just anyone connecting to your
iperf server, don't forget to
kill the server side with a Control-C when you are finished making
measurements.

In addition to simple TCP testing, it can also manipulate various TCP
parameters, test UDP streams, use multicast or IPv6, and even use a
custom defined data stream for testing. Running with the defaults
should give you a good basic idea of how much data you can cram
through your connection, particularly if it is not being used by any
other clients. For more complete details on some of the magic this
flexible tool can work, see the online documentation at http://dast.nlanr.net/Projects/Iperf/iperfdocs_1.7.0l.


/ 158