Architecture and Basic Principles for Data Administrators

The Architecture of SAS IT Resource Management

SAS IT Resource Management uses data integration and business intelligence components that are provided by the SAS Intelligence Platform as well as additional software that was designed specifically for the processing of IT resource data. At installation time, users can deploy these components across multiple tiers. Using a multi-tier architecture to separate major software functions supports flexible processing schemes. The following functions can be supported on the tiers that SAS IT Resource Management uses:

The client tier provides the interface between the user and SAS IT Resource Management. The client software enables you to perform data administration tasks, build reports, and view reports.
The middle tier provides server support for Java applications. The Web-based Gallery Manager application runs on this tier.
The server tier provides services that access data and perform background processing such as managing the metadata, executing stored processes, and performing resource aggregation calculations. In addition, the server tier provides the metadata for the files and tables that are input to and generated by the SAS IT Resource Management software.

With multi-tier processing, the tiers can be separated so that the processing can be done on multiple machines. The SAS IT Resource Management client must run on Windows. The SAS IT Resource Management middle tier and server can run in Windows, UNIX, or z/OS operating environments. Therefore, those software components for the server and the middle tiers can be installed on different machines according to the site requirements of an enterprise and the platforms that are supported by SAS for each tier of the SAS software architecture. For example, in a Windows environment, all of the processes that are needed by SAS IT Resource Management can be installed on a single PC. Other sites might prefer to install the client applications and middle tier on a PC and install the remaining components on the mainframe. Another option is to install the components needed for Web-based functions on a UNIX processor. The SAS representative can discuss these options and ensure the most appropriate configuration for a site.

The following diagram displays the architectural components that constitute the clients, middle tier, and servers that are used by SAS IT Resource Management.

Tiered Architecture of SAS IT Resource Management

Note: The multiple tiers shown in this diagram represent categories of software that perform similar functions. They do not necessarily represent separate machines or processors. In addition, your site might not need to use all of these software components. For example, if your site does not use the SAS Add-In for Microsoft Office, you do not need to install it.

The Basic Principles and Components of SAS IT Resource Management

Preliminary Tasks

SAS IT Resource Management has the flexibility to support customers whose usage of the product can range from the basic to the highly complex. Information consumers can work with specialized tools to access their IT data directly and generate ad hoc reports. Information consumers can use these reports to answer specific questions about the health and efficiency of the enterprise's IT resources. In all cases, the raw data for the resources that are being evaluated must be staged and aggregated. The resulting tables are input to the information maps and the reporting processes.

The following information describes the preliminary tasks that must be performed in order to use SAS IT Resource Management. After SAS IT Resource Management is installed and before the staging and aggregating processes can run, the data administrator uses SAS Management Console to perform the following setup tasks:

Connect to the metadata server.
Define new workspace servers.
Add users and groups to the online SAS Metadata Repository.

Note: The installation process for SAS IT Resource Management provides the definitions that are needed for the supplied groups. No definitions are supplied for users, so the data administrator must add these definitions.
Modify the default Access Control Template (ACT) for users or groups.
Define SAS Content Server mapping for use by SAS IT Resource Management report packages. The SAS Content Server requires an HTTP server in order to be used by SAS Web Report Studio and other SAS Business Intelligence Platform applications. This server must be configured so that its content and its metadata are synchronized and are compatible with the file structure of SAS IT Resource Management.
Define deployment locations for deploying jobs from SAS Data Integration Studio.

For more information about these tasks, see Working with SAS Management Console.

Using SAS IT Resource Management

After the preliminary tasks and initial setup are complete, you can use SAS IT Resource Management to process and evaluate IT resource data. SAS IT Resource Management provides the processes and the supporting technology that are required to regularly collect, aggregate, analyze, and report on the IT performance evaluation data that is vital to the health of an enterprise.

To start SAS IT Resource Management from the Windows Start menu, select Programs

SAS

SAS IT Resource Management.

In general, the processes that gather and monitor IT performance evaluation data consist of the following steps.

Collect the raw data about a resource.

IT performance data is information about IT resources, such as hardware, operating system software, networks, Web servers, databases, and applications. Raw (or unprocessed) data about the usage or performance of these resources is generated by the logging mechanisms that are inherent to IT resources or is created by the Enterprise Systems Management tools that are used to manage the IT infrastructure. The raw data is input to SAS IT Resource Management transformations that are supplied with SAS IT Resource Management software.

For information about how to generate and reference the raw data for supported adapters, see the various "Accessing Raw Data" topics in Data Sources Supported by SAS IT Resource Management Adapters.
Process (stage) the raw data.

The raw data is processed (or staged) by adapters that are tailored to the data sources that they process. The adapter's staging code performs functions such as normalizing measurement units, generating computed columns from the raw data, and checking for duplicate data. Staging is performed by transformations that are set up in jobs. The job that actually executes the transformation can run interactively but is normally scheduled to run in batch mode, depending on the requirements of your enterprise.

A unique staging transformation is supplied for each adapter that is supported by SAS IT Resource Management. It contains the code and can access the associated templates that are needed to process and load the raw data into staged tables. The staged tables can then be used as input to an Aggregation transformation or other SAS Data Integration Studio and SAS IT Resource Management transformations.
Information about these staging components is available in the following topics.
- For a list of the adapters that are supported by SAS IT Resource Management, see Supported Adapters.
- For more information about user-written staging code, see User-Written Staging Transformations.
- For more information about the staging transformations that are available for the supported adapters, see About Staging the Data.
Aggregate the data in the staged tables.
After the raw data is staged, it can be input to the aggregation step. An aggregation is the act or process of grouping data, using an operation that produces a statistic such as a sum, average, minimum, or maximum. The term aggregation can also refer to the grouped data that results from such an operation. Aggregation transformations can generate summarized aggregated tables or simple aggregated tables.
- Simple aggregations do not undergo any summarizations; they simply append new data as it was read by the staging transformation to an existing aggregation table. (These tables are typically called detail aggregation tables.)
- Summarized aggregations read data from a staged or user-defined table and then categorize and aggregate data according to the specifications of the Aggregation transformation.
Jobs that contain Aggregation transformations are created and updated in SAS Data Integration Studio.
An Aggregation transformation can contain multiple aggregations that each generate an aggregation table of data. SAS IT Resource Management provides the ability to customize Aggregation transformations based on a site's requirements. An Aggregation transformation can be customized to perform aging and filtering of the aggregated data. It can perform calculations that create additional columns of data and join columns from different tables. It can support summarizations of data into any choice of time periods. The Aggregation transformation can also compute statistics on the data, calculate percent of change of a statistic over a specified period of time, and calculate the rank of a statistic or a class or ID column.

An unlimited number of aggregations can be created for each staged table. For example, the same data source can be input to both a daily aggregation and a weekly aggregation within the same or separate Aggregation transformations. For more information about working with Aggregation transformations and aggregation tables, see About Aggregation Tables.

After the performance data has been aggregated, it is ready for the reporting processes.
Generate information maps.

For a supported adapter, SAS IT Resource Management generates transformations that create information maps that reference the tables of data that are generated by that adapter's Aggregation transformations. Information maps provide clearly labeled references for all data fields that are used to create and view reports. Information maps can be used in SAS Enterprise Guide and SAS Intelligence Platform applications such as SAS Web Report Studio in order to generate reports that provide domain intelligence about the adapters, such as CPU utilization, threshold analysis, and peak period analysis.

For more information about working with Information Map transformations to create and customize information maps, see About Information Maps.
Define and generate the reports.

SAS IT Resource Management generates tabular and graphical reports from report definitions that are supplied with the software, or created in SAS Enterprise Guide. Report definitions can be accessed by the Gallery transformation and, when the job that contains that Gallery transformation is run, the reports are generated and stored on the SAS Content Server. Report jobs are usually run in batch mode.

Report definitions can be run directly from SAS Enterprise Guide, and the resulting reports can be viewed interactively from the SAS Enterprise Guide Project window or from a Web browser. The reports can also be published to your Web server.
View the reports.

Reports that are generated by running the Gallery transformation are stored on the SAS Content Server where they can be accessed and managed by Gallery Manager. Gallery Manager enables information consumers to organize their reports into galleries and to set accessibility controls on those galleries. Galleries can be marked as private (for an individual's use) or public (for use by all Gallery Manager users.) Galleries can also be filtered so that they display only a subset of the available reports.

Reports can also be viewed on the Web through SAS Information Delivery Portal.

For more information about the reporting capabilities that are available using SAS IT Resource Management data, see SAS IT Resource Management: Reporting Guide.

Information consumers and performance analysts can also use these additional processes:

Generate and view ad hoc reports.

SAS Web Report Studio is a Web-based interface that can use information maps to select and report on data. Information maps are generated by Information Map transformations.
Use Microsoft Office products.

SAS Add-In for Microsoft Office enables SAS functionality to be accessed directly from the menus and toolbars of Microsoft Office word processing and spreadsheet products.

Performance analysts and information consumers who frequently work with Microsoft Office products such as Microsoft Word and Microsoft Excel can use SAS to access, analyze, and report on IT performance data that is available in information maps, and then distribute the results to other people in the enterprise.