Monitoring and Tuning Your Server

Analyzing Connection Data

By monitoring the number of connections, you can identify patterns of client demand for your server. Divide your PerfMon logs into regular time intervals, and look at the number of connections served during each interval. Observe the length of the processor queue and the processor use on each processor during periods of small, moderate, and large numbers of connections. This data shows how your configuration responds to each load level.

You can identify a processor bottleneck during a given interval by:

A long, sustained processor queue (more than two threads).
High use rates on one or more processors.
A curve in the graph of the Current Connections counter on any IIS 5.0 service performance object that reaches a high value and then forms a plateau. This pattern often indicates that additional connections are being blocked or rejected.

To prevent processor bottlenecks, make certain that a lengthy processor queue isn’t forming when you serve large numbers of connections. You can usually avoid a bottleneck during peak time by setting the connection limit to twice the average value of Current Connections. If the processor regularly becomes a bottleneck when servicing large numbers of connections, you might consider upgrading or adding processors, or limiting the maximum number of connections on the server. Limiting connections can cause the server to block or reject connections, but it helps to ensure that accepted connections are processed promptly.

Note The Active Server Pages, Web Service, FTP Service, and SMTP Server counters collect data at the Open Systems Interconnectivity (OSI) Application Layer. If any connections were blocked, rejected, or reset between the Transport and Application layers, counts of TCP/IP connections may not equal the sum of HTTP, FTP, and SMTP connections. For information about monitoring connections at lower layers, see Network I/O.