![]() | ![]() | ![]() |
The following topics provide some guidance on how you can optimize performance for your deployment of SAS Web Report Studio.
Hot fixes and maintenance release often include changes designed to improve performance. So, be sure to keep up to date with the latest version of SAS Web Report Studio.
For information about how to acquire the latest maintenance release and product upgrades, see Maintenance Releases and Product Upgrades .
For hot fixes for version 4.3, see SAS Web Report Studio 4.3: Hot Fix Downloads.
For hot fixes for version 4.2_M2, see SAS Web Report Studio 4.2_M2: Hot Fix Downloads.
For hot fixes for version 4.2_M1, see SAS Web Report Studio 4.2_M1: Hot Fix Downloads.
For hot fixes for version 4.2, see SAS Web Report Studio 4.2: Hot Fix Downloads.
For hot fixes for version 3.1, see SAS Web Report Studio 3.1: Hot Fix Downloads.
Recommendations can be found in the primary documentation and in usage notes. The following table lists some suggestions.
| Tip | Version 3.1 (SAS® 9.1.3) | Versions 4.2 and 4.3 (SAS 9.2) | Version 4.31 (SAS 9.3) |
|---|---|---|---|
| Use pooling for your workspace server. Beginning with version 4.2, a pooled workspace server is used by default. The result of this configuration is that the users of these Web applications will see much better performance than they would if they were connecting to a standard workspace server. You can also use client-side pooling. In version 3.1, you can set up pooling manually. | Convert a Workspace Server to Pooling | Understanding Server Pooling | Understanding Server Pooling |
| If you are using a standard workspace server, you can modify the workspace server startup options to specify a work library, a buffer size for writing files to the work area, and a limit on SAS memory usage. | Changing a Workspace Server's Launch Command | Workspace Server Configuration Tasks | Workspace Server Configuration Tasks |
| Set up your J2EE application server to use the correct Java Virtual Machine options, create a cluster of servers, tune your Web application server, and use an HTTP server to handle requests for static pages. Instructions includes sample deployment scenarios. | Best Practices for Configuring Your Middle Tier | Best Practices for Configuring Your Middle Tier | Best Practices for Configuring Your Middle Tier |
| Use the query cache, which is enabled by default. For reports that contain more than one data-driven object, this cache maximizes efficiency. The query cache builds a temporary common data table that can fulfill the needs of all data-driven objects in the report. When the query cache is used, complex queries that include functions such as joins and filters are run only once (to build the common data table). Each data-driven object in the report can then run simple extraction queries against the common data table. | Using the Query Cache | Using the Query Cache | Using the Query Cache |
There are ways to optimize your relational information maps, improve report design, and manage report handling Following are some tips for these areas.
| Tip | Version 3.1 (SAS® 9.1.3) | Versions 4.2 and 4.3 (SAS 9.2) | Version 4.31 (SAS 9.3) | |
|---|---|---|---|---|
| Make appropriate use of pre-rendered reports, such as manually refreshed reports and batch reports. Use report scheduling to control when batch reports are generated. For example, you can schedule reports to be generated on a nightly, weekly, or monthly basis. | Scheduling and Distributing Pre-generated Reports | Pre-generated Reports From SAS Web Report Studio | Pre-generated Reports From SAS Web Report Studio | |
| The structure of your data plus report design can impact the maximum tuple size, and, as a result, the memory usage and performance of a query. This usage note illustrates how data structure and report design affect memory usage, and then offers a suggestion for how to create a more efficient query. | Calculating the maximum tuple size for rendering relational data in crosstabulation tables in SAS Web Report Studio and SAS Web Report Viewer | |||
| Make sure that measure data items are defined to take advantage of an optimization feature introduced in the first maintenance release for SAS Web Report Studio 4.2. | not applicable | Optimize Measure Data Items for SAS Web Report Studio | ||
The following resources can help you ensure that queries submitted to third-party databases are optimized.
By default, the PROC SQL code generated for SAS Web Report Studio inserts the STRIP function to remove leading and trailing blanks on character fields on a filter expression. This default behavior can affect performance when the underlying database is not SAS. For more information, see Removing the STRIP function from filters in SAS Web Report Studio.
Add SQL_FUNCTIONS=ALL to your LIBNAME statement in SAS Management Console. Open the properties for the library and select Options ► Advanced Options ► Options. Then enter SQL_FUNCTIONS=ALL.
Note: Use of this option can cause unexpected results, especially if used for NULL processing and date, time, and timestamp handling. Read the documentation before using.
Using SQL Pass Thru to Improve Web Report Studio Performance
Configuring SAS Web Report Studio Releases 4.2 and 4.3 and the SAS Scalable Performance Data Server
Use the SQLREDUCEPUT System Option. For example, you might edit the sasv9_usermods.cfg file for the SASApp - Pooled Workspace server to include -SQLREDUCEPUT=dbms. Before using this option, read the documentation.
Some customers have reported performance degradation when the General Purpose Log becomes very large. The method for managing the log file depends on which version of SAS Web Report Studio you are running.
If you are running version 4.2 or 4.3, use one of the following samples to set a rollover mechanism for the SASWebReportStudio4.x.log file.
If you are running version 3.1, stop your Web application server, rename or delete your current WebReportStudio.log file, and restart your Web application server. The log file is typically located at \Lev1\web\Deployments\logs\WebReportStudio\WebReportStudio.log
If your reports are OLAP-based, then you can often improve performance by tuning your SAS® OLAP Server. Recommendations can be found in the primary documentation and in usage notes. The following table lists some suggestions.
| Tip | Version 9.1.3 sp4 | Version 9.2 | Version 9.3 |
|---|---|---|---|
| When you build cubes, you can set various options that improve and optimize cube creation and query performance. These options can be set for all aggregations in a cube or for a specific aggregation. | Specifying Tuning and Performance Options in Cube Aggregations | Specifying Tuning and Performance Options in Cube Aggregations | Specifying Tuning and Performance Options in Cube Aggregations |
| SAS Management Console enables you to optimize the query response time of SAS OLAP Servers by controlling factors such as the amount of cached metadata and aggregations, and number and configuration of threads used for queries. | Optimize SAS OLAP Servers | Tuning SAS OLAP Servers with Advanced Server Options | Tuning SAS OLAP Servers with Advanced Server Options |
| The performance of a query that references a non-default hierarchy might not be optimal when using clients such as SAS Web Report Studio, SAS® Information Delivery Portal, or SAS® Information Map Studio. One way to optimize these types of queries is to define a custom data item in the information map that references the top-level member of your hierarchy. | Usage Note 31698: Optimizing the performance of queries using non-default hierarchies in SAS® Java clients | ||
| When using NUNIQUE measures in your cube, it is very important to have a well-tuned cube. | Usage Note 35905: Tuning SAS OLAP cubes with NUNIQUE measures | ||
| If you are running the third maintenance release for SAS 9.2 (TS2M3), then install hot fix B25168. | not applicable | Usage Note 44742: General performance improvements for SAS® OLAP cubes are included with Hot Fix B25168 for the third maintenance release for SAS 9.2 (TS2M3) | not applicable |
| Take advantage of the SECURITY_SUBSET feature. The benefit to using the Security Subset feature is that the exclusion of the secured members is handled by the server directly, improving the performance of the response | not applicable | Using the SECURITY_SUBSET feature of SAS OLAP Server | |
SAS 9.2 Web Applications: Tuning for Performance and Scalability
Paper 217-30: Configuring J2EE Application Servers for Use with the SAS®9 BI Platform
| Product Family | Product | System | Product Release | SAS Release | ||
| Reported | Fixed* | Reported | Fixed* | |||
| SAS System | SAS Web Report Studio | Microsoft Windows 2000 Advanced Server | 3.1 | 9.1 TS1M3 SP4 | ||
| Microsoft Windows 2000 Datacenter Server | 3.1 | 9.1 TS1M3 SP4 | ||||
| Microsoft Windows 2000 Server | 3.1 | 9.1 TS1M3 SP4 | ||||
| Microsoft Windows 2000 Professional | 3.1 | 9.1 TS1M3 SP4 | ||||
| Microsoft Windows NT Workstation | 3.1 | 9.1 TS1M3 SP4 | ||||
| Microsoft Windows Server 2003 Datacenter Edition | 3.1 | 9.1 TS1M3 SP4 | ||||
| Microsoft Windows Server 2003 Enterprise Edition | 3.1 | 9.1 TS1M3 SP4 | ||||
| Microsoft Windows Server 2003 Standard Edition | 3.1 | 9.1 TS1M3 SP4 | ||||
| Microsoft Windows XP Professional | 3.1 | 9.1 TS1M3 SP4 | ||||
| Windows Vista | 3.1 | 9.1 TS1M3 SP4 | ||||
| 64-bit Enabled AIX | 3.1 | 9.1 TS1M3 SP4 | ||||
| 64-bit Enabled Solaris | 3.1 | 9.1 TS1M3 SP4 | ||||
| HP-UX IPF | 3.1 | 9.1 TS1M3 SP4 | ||||
| Type: | Usage Note |
| Priority: | |
| Topic: | Query and Reporting ==> OLAP (Online Analytical Processing) Query and Reporting ==> Interactive Querying System Administration ==> Servers ==> Web System Administration ==> Servers ==> OLAP System Administration ==> Performance Third Party ==> Products ==> Web Application Servers Third Party ==> Products ==> Web Application Servers ==> Jboss Third Party ==> Products ==> Web Application Servers ==> Tomcat Third Party ==> Products ==> Web Application Servers ==> WebLogic Third Party ==> Products ==> Web Application Servers ==> WebSphere System Administration ==> Web Query and Reporting |
| Date Modified: | 2011-03-28 15:34:56 |
| Date Created: | 2010-03-17 15:47:57 |



