Scope of Performance Data

In general, performance monitoring concentrates on how the operating system and any applications or services that are instrumented for performance-data collection use the resources of the system, such as the disks, memory, processors, and network components. Throughput, queue, and response time are terms that describe resource usage.

Throughput Defined

Throughput is a measure of the work done in a unit of time, typically evaluated from the server side in a client/server environment. Throughput tends to increase as the load increases up to a peak level. It then begins to fall, and a queue might develop. Throughput in an end-to-end system, such as client/server, depends on how each component performs. The slowest point in the system sets the throughput rate for the system as a whole. Often this slow point is referred to as a bottleneck. Performance monitoring tells you where bottlenecks occur in your system. The resource that shows the highest use is often the bottleneck, but not always—it can also mean a resource is successfully handling a lot of activity. As long as no queues develop, there is no bottleneck. Microsoft® Windows® 2000 reports throughput data on resources such as disks and network components.

Queue Defined

A queue can form under a few different circumstances. For example, a queue can develop when requests come in for service by the resource at a faster rate than the resource's throughput, or if requests demand differing, particularly longer, amounts of time from the resource. A queue can also form if requests occur at random intervals—for example, in large batches for a time and then none at all. When a queue becomes long, work is not being handled efficiently, and you might experience delays in response time. Windows 2000 reports queue development on disks, processors, and server work queues, tracking server message block (SMB) calls of the Server service.

Response Time Defined

Response time is the measure of time required to do work from start to finish. In a client/server environment, you typically measure response time on the client side. Response time generally increases as the load increases. You can measure response time by dividing the queue length for the resource over the resource throughput. As an alternative, the new trace log feature in the Windows 2000 performance tools allows you to track units of work from start to finish in order to determine response times.

The following sections describe how performance monitoring tools enable users to collect data about the throughput, queue formation, and response time of different system resources.

© 1985-2000 Microsoft Corporation. All rights reserved.