Page History
...
Table 2-4 Response time countersCounter Description
| Counter | Description |
|---|---|
| Client |
...
| time | If working in a LAN, client time should be a very small portion of total SAP response time. |
| Queue |
...
| time | Should be a very small portion (less than 1%) of the total SAP response time. When this value is high, other long-running transactions or batch programs prohibit proper throughput, or not enough work processes are available. |
| Application |
...
| time | Includes load time and should be no more than 60% of total SAP response time. When this value is high, the CPU is slowed down, there are SAP memory shortages or the program takes an excessive amount of time for load, queue or ABAP/4 application processing. |
| DB |
...
| time | Should be approximately 40% of total SAP response time. |
About application time counters
Application time counters help you determine which application time component constitutes a performance bottleneck. The following table describes the information displayed by the application time counters.
Table 2-5 Application time counters
| Counter |
|---|
...
| Description |
|---|
| Load |
...
| time | Should be no more than 10% of the application’s total response time. If this value exceeds 10%, this indicates that the SAP System is spending too much time loading the application. Check the Memory Resources counter to determine if there are any SAP memory shortages. If it appears that the CPU or memory resources are sufficient for the application’s processing, check for the following actions: |
...
|
...
|
...
|
| Enqueue |
...
| time | Should constitute a small percentage of the application time. |
...
When enqueue time is high, check or perform the following actions: |
...
|
...
|
| Roll-wait |
...
| time | The number of seconds that may be required to wait during roll-in of context information for dialog steps. The number of seconds required to transfer GUI control-related information to the front-end solution is also included. During processing, a program may result in a roll-out while waiting for other processing to take place. The process of roll-out and roll-in (in this case) is usually a change of memory pointers, and roll-wait time is minimal. In SAP 4.6 and higher, several communication steps, called round trips, can occur between the application servers and the front-end system. During the round trip, the application server transfers GUI control-related information to the front-end system. The work process is rolled-out and this time is recorded under roll-wait time. Network and front-end time are included as part of roll-wait time, but only as it applies to the control-related information transferred between the application server and the front-end system during the execution of the dialog step. Network time associated with the first request and the last request is therefore not included in roll-wait time. Roll-wait time should be very low. When high, perform the following steps: |
...
|
...
|
| Process |
...
| time | Should constitute 50-60% of the application’s total SAP response time. When this percentage is higher than it should be, system or SAP memory resources are an issue or the application is spending too much time in application logic. Take the following steps: |
...
|
...
|
...
|
About DB time counters
DB time counters identify whether database response time is attributed to sequential reads, direct reads or changes by comparing their times.
DB time counters refer to the Time Per Request section to determine if response times fall within acceptable performance ratings.
When database time is attributed to sequential reads or direct reads, see the Application Buffer Ratio and investigate whether buffering techniques are optimized for read requests.
The following table describes the information displayed by the DB counters.
Table 2-6 DB time counters
| Counter |
|---|
...
| Description |
|---|
| Sequential |
...
| Reads | The response time for logical database requests associated with sequential reads (for example, "select * from ..." ABAP/4 statements). |
| Direct |
...
| Reads | The response time for logical database requests associated with direct reads (for example, "select single ..." ABAP/4 statements). |
| Changes |
...
| The response time for logical database requests associated with database inserts, updates and deletes. |
About requests counters
Requests counters show the volume of database read requests and change requests. The following table describes the information displayed by the Requests counters.
Table 2-7 Requests counters
| Counter |
|---|
...
| Description |
|---|
| Sequential |
...
| Reads | The number of logical database requests associated with sequential reads (for example, "select |
...
| * from ..." ABAP/4 statements). |
| Direct |
...
| Reads | The number of logical database requests associated with direct reads (for example, "select single ..." ABAP/4 statements). |
| Changes |
...
| The number of logical database requests associated with database inserts, updates and deletes. |
See “About DB time counters” on page 24 .
See and “About average server buffer ratio” on page 25.
About time per request counters
Time per request counters identify when database requests do not fall within their performance boundaries. The following table describes the information displayed by the Times per request counters.
Table 2-8 Time per request countersCounter Description
| Counter | Description |
|---|---|
| Sequential |
...
| Reads | The response time, per request, for logical database requests associated with sequential reads (for example, "select * from ..." ABAP/4 statements). |
| Direct |
...
| Reads | The response time, per request, for logical database requests associated with direct reads |
...
| (for example, "select single ..." ABAP/4 statements). |
...
| Changes | The response time, per request, for logical database requests associated with database inserts, updates and deletes. |
When analyzing these counters, keep in mind the following issues:■ Sequential
- Sequential reads should be less than 40 milliseconds per request.
...
- Direct read requests should be less than 10 milliseconds per request.
...
- Change requests are expected to be greater than 25 milliseconds per request.
...
- When per request times are not within their expected performance boundaries:
...
- Use the DB Time section to determine the importance of the types of read or change requests in question, with respect to the overall response time.
...
- Use the Requests section to understand the request volume involved in read and change requests.
...
- When sequential reads or direct reads are performing poorly, see the Application Server Buffer Ratio section to investigate whether buffering techniques are optimized for read requests.
...
- When changes or reads are performing poorly and your Application Server Buffer ratio appears normal, but the number of returned rows appears unacceptable, the database is likely the source of the problem and should be examined for additional indexes or other causes.
About average server buffer ratio
The average server buffer ratio illustrates the degree of optimization related to application server buffering techniques, which optimize performance for database requests by avoiding the need to transfer read requests to the database server.
The following table describes the information displayed by the average server buffer ratio.
Table 2-9 Average server buffer ratioCounter Description
| Counter | Description |
|---|---|
| Sequential |
...
| Reads | The percentage of time sequential read requests were handled by the application server, instead of the database server. Sequential read requests refer to logical database requests associated with sequential reads (for example, "select * from ..." ABAP/4 statements). |
| Direct |
...
| Reads | The percentage of time direct read requests were handled by the application server, instead of the database server. Direct read requests refer to logical database requests associated with direct reads (for example, "select single ..." ABAP/4 statements). |
Keep in mind the following issues when analyzing these counters:■ App
- App Server Buffer Ratios should typically be 90% or greater. The higher the percentage indicates that fewer read requests are transferred to the database server and performance is optimized.
...
- If this percentage is low, determine the importance of the App Server Buffer Ratio to your transaction's database performance:
...
- Check the DB Time section of this view to determine if database response time is mostly attributed to reads
...
- (sequential and direct) or changes.
...
- Database time attributed mostly to changes suggests that the Buffer Ratio is not an important factor to overall response time—this is because changes are always transferred to the database server.
...
- When database time is influenced by reads, the buffer ratio is likely to impact overall database time. Buffer usage, for example, can reduce response time for database reads by as much as 10 times.
...
- Consider implementing 1 of these 3 buffering techniques to increase read performance and reduce load on the database server.
...
- Buffer data as context in the shared memory buffer.
...
- Buffer data in a table using a generic key or single record buffer.
...
- Buffer data using an internal table of a function module.
...
- Note the following regarding buffering techniques:
...
- Restrict buffering techniques to read-only data as much as possible to minimize buffer synchronization across application servers.
...
- When selecting a buffering technique, take data size requirements into consideration. An instance typically has about 50 MB of data in the shared memory and buffers, whereas, information stored in internal tables should not exceed 5 MB.
...
- Buffered tables take much more effort to change than non-buffered tables. Therefore, restrict buffered tables to read-only data stored in customizing tables or small tables.
About memory resources
Memory resources determine if an application is receiving a sufficient amount of SAP application server memory resources.
The following table shows which counters are included in the Memory resources.
Table 2-10 Memory resources.Counter Description
| Counter | Description |
|---|---|
| Private Mode | Private |
...
mode conditions should not occur. When private mode conditions occur, the SAP application server's shared memory is insufficient to meet the demands of the application, and the following occurs: |
...
|
...
|
...
|
| Extended |
...
| Memory | To determine memory shortage impact on an application's response time, do the following: |
...
|
...
|
...
|
...
|
...
|
| Anchor | ||||
|---|---|---|---|---|
|
Precise is a Web-based application. You can access the Precise user interface using the Internet Explorer browser, version 6.0, or later. The syntax of the Precise URL address is http://<server>:<port>,where <server> refers to the Precise FocalPoint server and <port> refers to the port number used by the GUI Web server. By default, the port number is 20790. For example: http://beanotown:20790.
This URL provides secure access to the StartPoint using authorized roles. From here, you can launch all Precise products. It gives you a quick overview of the status of your environments and access to the AdminPoint, where you can perform various management tasks (see the Precise Administration Guide for details).
| Info |
|---|
...
You must have local administrator privileges on the server where the StartPoint is running. |
To launch your product using StartPoint1 Type
- Type the address of the StartPoint user interface into the Address bar of your browser and click Enter.
...
- The Precise login page opens. The login page provides secure access to Precise and to your specific product.
...
- Specify your authorized role name and password. By default, both role name and password are admin.
...
- For more information about role names, see the Precise Administration Guide.
...
- Click Login. The StartPoint page opens. This is the Precise home page.
...
- On the Product Selection bar, from the drop-down list, select the product you want to launch.
Precise. Performance intelligence from click to storage. Learn more > >
...