Page History
This section includes the following topics:
- About the Statistics workspace
- How the Statistics workspace is structured
- About the entities you can examine in the Statistics workspace
- How to identify performance problems
| Anchor | ||||
|---|---|---|---|---|
|
The Statistics workspace displays statistical information on all .NET instances. The workspace can be used to monitor your system's current state as well as historical statistical information.
...
Use the Statistics workspace to periodically examine the health of your system. Alternatively, you can use the Statistics workspace to fully analyze a performance problem. For example, you can use the overtime graph for a specific counter to get a wide view and examine the counter behavior in the suspicious instance and time slice.
| Anchor | ||||
|---|---|---|---|---|
|
The Statistics workspace displays information on a selected entity and its associated entities. For example, it is possible to associate with all counters that are related to a specific instance, by selecting the Counters entity from the Association controls.
...
The selected entity is always reflected in the workspace heading, which serves as a point of orientation. The highest-level entity you can view information for in the Statistics workspace is the Tier. You can view information on an instance by selecting it from the Instance list.
...
From the Association area, you can also drill down to another entity by clicking a row in the table. A drilldown drill-down affects the whole workspace. When you drill down to another entity, the workspace heading displays the new selection; the Main area displays information on the newly selected entity, and the Association area displays the entities associated with the selected entity.
...
See “How most tabs are structured” on page 9.
| Anchor | ||||
|---|---|---|---|---|
|
The Statistics workspace displays information on different entities. This section provides an overview of all entities, their meaning, and their views.
The following entities are available:
- Instances
- Counters
...
| Info |
|---|
For an explanation of the counters displayed in the Statistics workspace, move your pointer over the counter name in the table to view a ToolTip, or see the Microsoft performance monitor tool. |
About the Counter entity in the Statistics workspace
...
Table 5-1 Counter information in the Association area
Column Description
...
| Column | Description |
|---|---|
| Category | Displays the perform category of the counter. |
| Counter |
...
| Name | Displays the unique counter name. |
| Counter |
...
| Value | Displays the sampled counter value. |
| Anchor | ||||
|---|---|---|---|---|
|
To determine whether your .NET application is performing optimally it is necessary to analyze performance measurements over time. The Statistics workspace provides information about many Microsoft performance counters related to .NET and ASP.NET instances, as well as some server related counters.
They are grouped into several predefined views and categories that enable you to locate performance problems such as CPU, GC, and memory usage.
Below is a list of common performance problem areas. Each section will refer to important counters to watch using the Statistics workspace when analyzing your instance performance.
How to examine ASP.NET-related statistics
You can examine your ASP.NET instance counters over time to confirm the ASP.NET normal operation over time. For example, you can find out if your ASP.NET instance experiences many failed requests. If so, it indicates a problem with your application or server that needs further investigation.
...
Counters can be examined by drilling down on a specific counter in the counters table area. Main counters:Request Counters Requests Total: shows
| Counter | Description |
|---|---|
| Request Counters | Requests Total. Shows the number of requests served by your ASP.NET instance (IIS). These includes all pages processed by the ASP.NET ISAPI filter - not just .aspx, but also .asmx, .axd, .ashx and others. Requests Failed |
...
. Shows the number of requests that failed. A high number might indicate an exception condition is thrown repeatedly in your application. Other Requests counters |
...
. Requests not authorized, Requests authorized, Requests succeeded, Requests Timed Out., Requests Current, Requests Rejected, Requests Queued, Requests Not Found. |
| Session |
...
| Counters | Session Abandoned, Sessions Timed out, Sessions Total. |
| Worker Process |
...
| Counters | Worker Process restarts and Worker Process Running counters can be used to detect situations where due to a failure your ASP.NET application restarts frequently. |
| ASP.NET Error |
...
| Counters | ASP.NET errors counters can be used to detect errors thrown by your web application and not handled by your code. Explanation about these counters can be found in the SmarTune workspace Advice tabs regarding these counters:
|
| Cache Usage |
...
| Counters | These counters can be used to detect bad usage of cache by your application. There are several ways to use caching in your ASP.NET application. Explanation about these counters can be found in the SmarTune workspace Advice tabs regarding these counters:
|
How to examine Garbage Collection related statistics
...
- In the Time Frame list, choose the period of time you want to analyze.
- In the Instance list, choose the .NET instance you want to analyze.
- On the Main area View controls, click Garbage Collection.
The % Time in GC (Avg) counter shows the time your application spends doing garbage collection. A high value indicates you are experiencing problems with your applications memory management scheme. This could occur for many reasons; the most common are:
...
- In the Time Frame list, choose the period of time you want to analyze.
- In the Instance list, choose the .NET instance you want to analyze.
- On the Main area View controls, click CPU & Memory.
The CPU Consumption (Avg) counter shows average CPU time consumed by your .NET application. Use this graph to find peaks of high CPU usage. High CPU usage can be caused by many things, including bad programming, too much time in GC, etc.
...