Introduction to Performance Monitor Strategies
What have performance monitor and sex education in common? In
both cases teaching the mechanics easily, but in both cases the emotional
element is much harder to get across. Whilst my analogy is imperfect, my
point is valid, mastering performance monitor is a higher level art form,
whereas, merely
adding counters is easy.
Performance Monitor Topics
♦
To get the most from Performance Monitoring, you need a clear
goal
To discover a bottleneck, performance monitor will unearth areas
of high demand, for example, RAM, Processor, or Disk. Logging key counters
can also anticipate problems, for example imminent disk failure.
Baseline, investigating the impact of services on system resources.
What is the network like at night when no-one is collecting email or querying SQL?
Linked to baseline analysis is capacity planning. Do you know what would
happen if you put another 100 users on that subnet? No? Performance
monitor will let you play 'what if..' games.
Learning about systems, is an unexpected bonus, you cannot help but
learn how the operating system works when you measure its performance counters.
For example, insights that database servers use more memory than file and print
servers.
Maximising resources, at the very least, performance monitor will give
you ideas for load balancing servers. Analysis may also unearth
incorrectly configured resource, for example network cards set at 100 Mps instead
of 1000 Mps. Time spent monitoring existing servers will repay when you
are working out the specification for a new machine.
Testing, a double edged sword. Having a test network to try new
configurations will ultimately improve the production network. Performance
monitoring may be the catalyst to commission or strengthen a test network.
Another selling point is that a top specification test network can provide spare
machines when a server on the live network needs to be repaired.
Guy Recommends: The Orion Network Performance Monitor (NPM) 9.5
Orion's performance monitor is designed for detecting network outages.
This NPM will guide you
through troubleshooting by indicating whether the root cause is a broken link,
faulty equipment or resource overload. Because it produces
network-centric views, it is intuitive to navigate, and as result you can
see easily what's working and what's not.
Perhaps Orion's best feature is the way it suggests solutions. Moreover, if
problems arise out of the blue, then you can configure Orion NPM 9.5 to notify
members of your team what's changed and how to fix it.
If you are interested in testing a professional performance monitor on your
network, then I recommend that you take advantage of Solarwinds' offer of a
download a free trial of Orion's Network Performance Monitor.
When I was a young 'greenhorn' on my first job, my findings were initially
dismissed because they said that my results were skewed because performance
monitor itself was putting a load on the server. Well, I learnt from that
experience and subsequently, I always monitor the servers from my laptop or a
spare machine.
Another trap that I fell into was trying to monitor a SQL. The problem
was that there were no SQL specific counters on my laptop, what I had to do was
install SQL locally on the machine which was monitoring the live server.
Microsoft Network Monitor v3.2 is a tool which captures TCP/IP
packets and reveals their source and destination addresses along with
detailed information stored in the datagram header. Network Monitor 3.2 works on all modern Windows
operating systems, such as Server 2008, and Vista.
Troubleshooting connectivity problems.
Let
us imagine that DNS is not working. If you capture the appropriate
frames with the Network Monitor, you may discover from the destination
address that your machine is trying to connect to a non-existent DNS
server.
Calculating server response times.
Each
packet has date / time information, thus you can measure response times
for conversations between your computer and various servers. If
necessary you could instigate a conversation with ping.
TCP re-transmissions.
A significant number of re-transmissions could indicate an intermittent
connection problem.
Identify broadcast traffic.
Broadcast traffic is
an old enemy of network managers. You could use seeking
broadcast or multicast traffic as an opportunity learn more about Network
Monitor, while you check for a well-known network problem.
Guy Recommends: SolarWinds LANSurveyor
LANSurveyor will produce a neat diagram of your network topology. But that's
just the start;
LANSurveyor can
create an inventory of the hardware and software
of your machines and network devices. Other neat features include dynamic
update for when you add new devices to your network. I also love the ability to export
the diagrams
to Microsoft Visio.
Finally, Guy bets that if you take a free trial of LANSurveyor then you will
find a device on your network that you had forgotten about, or someone else
installed without you realizing!
Download a Free Trial of LANSurveyor
More Help for Detecting Computer Bottlenecks
|
Learn
the secrets of which counters to monitor. Master performance monitor
logging,
develop your skills with structured exercises and examples. Print
out a copy to read, while you design logs and alerts to detect network
bottlenecks.
