April 15, 2000
Understanding Application Statistics
Some applications can be a networking nightmare. A poorly written app can bring a network to it's knees, and others programs can randomly degrade the network at the most inopportune times. To help troubleshoot applications, a network analyzer can provide statistics and feedback at the application's level. Each application can have its own specialized statistics, unlike the finite number of statistics available at the physical and data link layers of the network.
Application Response Time
Unlike the response time measurements from a transport protocol such as TCP, the application response time is specific to the program's traffic between the server and the client. This statistic itself doesn't usually cause a protocol analyzer to report an error, although the percentage of slow responses to normal responses should be observed to determine the responsiveness of a server.
Denied Count
Client applications usually make a request to a server and expect a successful response. If the client's request is for a resource doesn't exist or the client doesn't have rights to the resource, the server denies the client's request. Some analysis tools include a customizable filter time to determine if a denied request should be counted. For instance, if the filter time is set to one second, the analyzer won't report denied requests unless more than one request is denied in that one second period.
Loop Percentage
A poorly designed application can cause the program to repeat a request to the server. A poorly designed application can cause the program to repeat a request to the server. A poorly designed application can cause the program to repeat a request to the server. You get the idea.
A repeated request is one that is continually sent, even though the proper response to the previous request has already been seen by the analyzer. Many analyzers can calculate the ratio of looped requests to normal requests to provide a percentage statistic.
Slow File Transfer
File transfers can be qualified as efficient if they send as much information as possible in the shortest amount of time. The key to this efficiency is the total amount of data in each packet of the transfer. The size of the data within the packets of a file transfer is observed by the analyzer, and a file transfer data size per packet that falls under a predefined threshold is marked as a slow file transfer. Network designers should use this statistic to help configure and maintain all devices along the transfer path to maximize the file transfer efficiency of the network.
Posted by james_messer at April 15, 2000 02:30 PM
Thanks for signing in, . Now you can comment. (sign out)
(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)