In order to test and
troubleshoot networks we need tools that allow us to generate
network traffic and analyze the network's throughput performance.
This is true for both wired and wireless networks, but we'll mostly
focus on 802.11 (WiFi) networks because analysis tools are
relatively scarce or, when available, tend to be rather
expensive and overly complex to use. Also, because WiFi networks are
sensitive to RF interference from other wireless devices, they are
more difficult to troubleshoot and transient changes in the local
environment may affect their performance. Tools that are typically
used to troubleshoot wireless networks report signal strengths of RF
interference or beacons from an access point in units of dBm or RSSI
(relative signal strength indication). But what do these really
mean? How do these translate in terms of the performance of your
wireless network?
Our wireless network adapter may
report the beacon strength of our access point to be -53 dBm or -65
dBm or -73 dBm or <fill-in some number>. Or, our RF spectrum
analyzer displays a -74 dBm interference peak in the middle of
channel 6. Or, we and our neighbor or the business next door have
our access points configured to use channel 11? Is this good?
Bad? Doesn't matter? Ultimately, the bottom line and what we care
about most is throughput -- that is, how many bytes-per-sec can be
transferred from one node on the network to another. The dBm and
RSSI numbers don't mean much if you can't somehow relate them to
a performance metric -- the most relevant being 'bytes / sec'. The point is before
we can really begin to troubleshoot a wireless network you need a
way to benchmark its performance, so as modifications are made
we can determine whether or not they really make a difference in
the network's performance.
This is where
NetStress fits in. It's a simple tool, originally
created for our internal business use, which we later realized would also
be of value to others. Here's the ideal scenario: You install a new
wireless network and it works perfectly. Run NetStress
and record the benchmark results for later reference. Sometime in
the future when either you or others suspect performance has
declined then run NetStress again and
compare the latest results with those you gathered when the network
was first installed and working "perfectly". The results of that
comparison will indicate whether or not there really is a problem
and dictate which steps to take next (if any). Regardless of your
technical expertise or how expensive your diagnostic tools, if you
are modifying a wireless network or making decisions to modify it
and you are not testing throughput then you risk wasting time and
resources
going down the wrong path.