10.1 Performance Metrics

The condition that occurs when a system is unable to keep up with the demands placed on it.

• Bottleneck

• Latency

• Bandwidth

• Throughput

The amount of data that could be transferred from one place to another in a specific amount of time.

• Bottleneck

• Latency

• Bandwidth

• Throughput

The speed that data packets travel from source to destination and back. All packets experience some level of this.

• Bottleneck

• Latency

• Bandwidth

• Throughput

The amount of data that is transferred from one place to another in a specific amount of time.

• Bottleneck

• Latency

• Bandwidth

• Throughput

To analyze this, you can include:
The number of protocols that are running on the network.

• The number of errors typically logged.

• The frequency of use of specific applications.

• Bandwidth used.

• The number of users logging in each day.

• Baseline

• Latency

• Bandwidth utilization

• Error rate

• 
  When packets sent together arrive at different times, they could arrive out of order. This causes what’s called a jitter. A packet that arrives after its expiration is dropped. This uses what metric?

• Baseline

• Latency

• Bandwidth utilization

• Error rate

• The percentage of available _______ that is being used. A network should not regularly utilize all its available _______. Systems perform best when additional _______ is available for usage spikes.

• Baseline

• Latency

• Bandwidth utilization

• Error rate

• 
  A calculation of how often bits are damaged in transit due to electromagnetic interference or other interference. If a packet is damaged too much, it is dropped

• Baseline

• Latency

• Bandwidth utilization

• Error rate

The percentage of time this spends doing non-idle tasks.

It should be relatively low, up to 40% on average.

It will spike, when a major task is launched.

If it is consistently high, then the CPU is likely the bottleneck.

• Processor utilization

• Hard disk utilization

• Memory utilization

• Temperature

The percentage of time that this subsystem is reading from and writing.

If this value is consistently over 90%, check the following statistics to identify the source of the activity.

Average ____ Queue Length

Memory

• Processor utilization

• Hard disk utilization

• Memory utilization

• Temperature