IBM Cognos Administration overview - IBM Cognos

Meeting service commitments to the business and anticipating and responding to changing business requirements—all within tight budget constraints—are fundamental IT tasks.

As noted in The Performance Manager:Proven Strategies for Tuning Information into Higher Business Performance (ISBN-10: 0973012412), IT also has a strategic role to play, which means going beyond the fundamentals. IT can be to your company what high-tech firms have been to the economy—a catalyst for change and an engine driving rapid growth. Why? Technology and informationhave become so important to how companies operate that even small changescan dramatically affect your ability to manage the performance of the business.

If IT spends all of its time and resources ensuring service commitments, it is impossible to initiate and support more strategic opportunities. In this way, anyway in which to reduce the time and effort to administer and maintain enterprise applications becomes a strategic gain.

IT publications often compare IT’s job and the managing of enterprise applications to conducting an orchestra. There are multiple moving parts (databases, servers, networks) that must all work together to provide a positive user experience. Working in concert, the application can become a critical part of the business. Out of tune, it can become an immediate distraction, a critical failure, or worse, something deemed untrustworthy by users. If this happens, IT has lost user buy-in and, most likely, a portion of its return on investment.

IBM Cognos Administration capabilities

IBM Cognos Administration provides capabilities for IT professionals to manage their business analytics systems proactively to prevent problems before they occur. These system administration capabilities let IT address the important considerations necessary to exceed commitments to the business while respecting budgetary and other resource constraints.

System administration capabilities

System administration capabilities

Understand usage patterns

Understanding usage patterns is important for both troubleshooting immediate system issues and for performance-tuning activities over the life cycle across the many components of the business intelligence system.

There is no one simple approach to understanding usage patterns. They are unique to the cycle of how the organization gathers information, reviews it, and distributes it. For example, managing a centrally located user community that distributes pre-generated reports over email is not the same as managing a group of users who are always on the road and who want to access all information with a mobile device.

Certain patterns are more obvious and better defined. For example, quarter-end and year-end typically generate extra system activity for most departments. Others vary depending on the industry or the culture of how the organization communicates.

Understanding usage patterns means knowing the number of users that access a system at any given period of time, where these users are located, how much time they spend using the solution, and how they use the solution.

To gain a better understanding of usage patterns, certain metrics provide a good indication of how well the business intelligence solution is adapting to how people use it. Whereas no one metric provides a complete picture of the system, these metric examples are a starting place for tracking usage patterns:

  • Number of sessions
    This system-level metric provides the number of user sessions actively logged onto the system. Whereas this does not tell you how the business is using the system, it does provide insight into how many people choose to use the system.
  • Number of processed requests
    This metric provides the number of requests received at a specific point in time. Looking at processed requests alone provides a sense of magnitude in terms of system usage. What are the peak periods versus the slower periods? How much are business intelligence applications being used? This is useful for determining optimal times to schedule batch reporting.
  • Number of queue requests
    This metric provides the number of requests that have gone through the queue. A high number of queue requests might indicate a high volume of system activity at a particular point in time or an issue that needs to be addressed. With a more consistent and deeper understanding of usage patterns, an IBM Cognos BI system administrator can determine whether this is regular activity or an anomaly that needs explanation and potential action.
  • Longest time in queue
    The time in queue high watermark metric provides the longest time that a request has been in the queue. As it increases, it indicates more system activity and longer waits. This is a useful metric to monitor for changes on a regular basis. If queue times increase over a short period, there might be a change to the usage patterns that requires further investigation.

Understand IBM Cognos BI system environment

IBM Cognos BI is built on a modern service-oriented architecture (SOA) platform. This flexible platform offers many deployment options based on the preferred IT infrastructure and enterprise architecture strategy. As the organization extends the IBM Cognos BI system, it should monitor the initial deployment strategy to ensure that it still fits the current deployment landscape.

For example, you might have initially deployed business intelligence in a centralized server environment with certain affinities in place to handle usage patterns for the initial deployment. You might have decided to dedicate a particular server to running reports that are processing intensive or you might have allocated a local server for a geographic location without adequate network access. As the solution expands, the assumptions driving the initial architecture infrastructure might change. You might need to revisit them to ensure optimal solution performance. To better understand the overall health of the business intelligence system environment, consider the following metrics:

  • Successful requests per minute This metric shows the average amount of successful requests relative to the amount of processing time it took to execute them. The algorithm for this metric was devised to provide a real measure that is not impacted by periods of inactivity. This metric is useful for determining and tracking server throughput.
  • Number of processed requests per dispatcher
    The number of processed requests per dispatcher is a good indicator of load balance in the business intelligence system. If one dispatcher is handling a heavier load, you need to understand why. Is this a deliberate configuration choice based on the usage patterns, or does it require further review?
  • Percentage of failed requests
    This metric provides the percentage of failed requests based on the total number of requests handled. This metric gives you trending information over longer periods of time to understand how the business intelligence services are performing.

Understand the business expectations

Whereas formal service level agreements (SLAs) provide a structured approach to communicate and set system expectations, IT needs to keep the communication channels open with business owners and ensure that priorities align with organizational strategy.

Business expectations set the agenda for what metrics to track, the thresholds to set, and how to prioritize follow-up actions. For example, a company’s strategy might center on customer service with an objective to improve call center performance. Ensuring that critical call center information is readily available on demand will be at the top of the priority list. IT would want to monitor and ensure system uptime and report response times related to the call centers.

By contrast, if an organization is trying to reduce costs through process optimization, then weekly or monthly reports are critical to manage. These reports might not demand faster response times, but monitoring failure rates would be key to ensuring that they are delivered in time to enable a streamlined process.

Determine what thresholds to address

Setting metrics and gathering data on usage patterns and technology environments and understanding business expectations, is important to effective system management. Identifying thresholds for those metrics simplifies IT’s task (and communication with business owners) by giving the context to determine when to take action. Thresholds make IT proactive. IT can flag issues before they affect users, breach SLAs, and lead to support calls.

There can be hundreds of system metrics. Taking time to understand what metrics are vital for system management of your business intelligence solution is essential before setting any thresholds. IT is no different from the users it supports. Too much information makes taking the right course of action as difficult to determine as having no information at all.

For example,if the longest time in queue starts increasing, it might require further investigation. If queue and wait times increase, users might wonder whether there is a system issue. Having an agreed-upon threshold on this metric would identify the point when IT needs to take further action to understand what is happening. Users would know the threshold and also know that their IT department is dealing with the issue.

Track and evolve over time

After IT has identified key metrics and set thresholds, it can respond to current situations proactively to avoid business disruption. The next consideration for system management is making system metrics work for IT and the business over the long term.

Metrics provide IT and business users with insight into changing usage patterns and technology environments over time. With this information, IT can tune the business intelligence solution and adapt metrics and thresholds to maintain, meet, and improve service standards.

To accomplish this, IT needs business intelligence reporting on its system information. As described in The Performance Manager, IT must use dashboards, scorecards, reports, analysis, and alerts to deliver the correct information to drive improved decisions within the IT department. Better IT decisions can affect everyone across the organization.

Summary: Five steps to effective business intelligence system management

To effectively manage their business intelligence solution, IT managers need to understand their business intelligence solution:

  • Usage patterns
  • Business intelligence system environment
  • User expectations

It also means being able to resolve problems and take actions quickly to prevent issues from occurring. To do this, IT needs to monitor metrics, set thresholds, and analyze usage patterns and key aspects of the business intelligence system environment. IT professionals can use this data to initiate conversations with their business partners—bridging technology and information requirements—to stay in touch with user expectations and service commitments.

IT has the facility to adapt to changes in usage and new business needs. This becomes the means to drive continuous improvement in the performance of your business intelligence solution, and the roadmap to more effective performance management.

With IBM Cognos BI, business intelligence administrators and IT professionals gain new facilities to manage the health of the business intelligence system. Task-oriented system monitoring gives administrators a new, consolidated view of all system activity, from scheduled and interactive reports to servers and dispatchers. Proactive administration through detailed system metrics and the ability to set thresholds that can be monitored let IT professionals identify and correct anomalies.

IBM Cognos BI, using the IBM Cognos Platform, delivers broad system management for IT to confidently deploy business intelligence, proactively manage the system, and respond to new requirements while meeting on-time service commitments.

With these system management capabilities, IT can deliver on its fundamental tasks (that is, meeting service-level commitments and responding to changing business requirements within budgetary constraints) and have the capacity to drive strategic objectives. In this way, IT can realize its full potential as leaders and change agents within organizations.

The IBM Cognos Administration user interface

IBM Cognos BI provides a centralized, web-based administration console that provides administrators with the tools that are necessary to manage the IBM Cognos application.

To launch the administration console, use one of the following methods:

  • Click the IBM Cognos Administration link from the Launch menu within IBM Cognos Connection.
  • Launching the administration console from IBM Cognos Connection
    Launching the administration console from IBM Cognos Connection

  • Select the Administer IBM Cognos content link from the welcome menu
  • Welcome to IBM Cognos software menu
    Welcome to IBM Cognos software menu

The administration console includes the following tabs :

  • Status
  • Security
  • Configuration
  • Index Search

Each tab contains a set of logically grouped functions, called tasks, on the left side of the user interface.

Administration console tabs
Administration console tabs

Status tab

The Status tab is designed to provide administrators with the visibility and insight into which objects are currently running both interactively and in the background (scheduled or batch), which have executed in the past, and those that have been slated to execute in the future. In addition to the object execution details, the Status tab provides a health check of the overall IBM Cognos environment.

Status tab and the associated tasks

Current activities task
Status tab and the associated tasks

The current activities task exposes all the objects in the environment (interactive or background) that are currently running in an executing, pending, waiting, or suspended state. Looking at the top of the Current Activities interface, a summary graph displays all of the current objects and the state that they are in. The default view displays what is currently being handled through the background (batch) processes. It is possible to toggle between background and interactive activities, but it is not possible to view both simultaneously.

  1. Locate the Filter frame in the lower-left side of the page.
  2. Toggle between the background activities and interactive activities by selecting the appropriate radio button.
  3. To change the display in the right part of the interface, click Apply.

Focusing first on the background activities, in addition to the summary chart, all of the objects currently in the system display in a paginated list. Looking at an entry, the following items are visible in the default view :

  • The object name
  • The time that it was requested
  • Who executed the object
  • The dispatcher and process ID handling the request
  • The last successful execution time
  • The status
  • The priority assigned to the object

Executing job in the current activities task
Executing job in the current activities task

Figure shows a job in an executing state that was requested by Sam Carter. To see more details about the object, click the Show Details icon on the toolbar.

This expands the list to include a second line of detail. Conversely, the Hide Details icon removes the additional line of detail. Keep in mind that the detailsthat are shown might differ from task to task due to the nature of the task and thelevel of information available to the task.To help reduce the number of objects displaying or to quickly isolate and focus inon particular objects, a series of filter options are included in the Filter section inthe lower-left frame.

List of filters available to help reduce the results displayed in the list
List of filters available to help reduce the results displayed in the list

The default filter options allow the result set to be filtered by user running the object, the current state that it is in, and the priority that has been assigned to the object. Clicking the Advanced options link allows for filtering by the owner of an object being executed, the type (job, report, agent, import, and so on), a dispatcher (which provides administrators with the insight to identify which dispatcher is handling which objects), and scope (package or folder in which the object resides). Any single filter option or a combination of options can be selected, and after the options have been determined, click Apply for the result set to be updated.

Moving on to the interactive activities, most of the level of detail that was available in the background activities view is still applicable, with a couple of minor exceptions:

  • The objects running interactively have fewer states. Either it is pending (queued) or is executing.
  • The only default filter option available for the interactive view is the ability to filter by status. The advanced options provide the ability to filter by dispatcher.

Past Activities task

As the name implies, this task displays objects that were executed in the past. One important thing to note about this interface is that only the objects that were executed in the background will be recorded. Interactive requests do not record histories that can be displayed through this task. For this type of information, you need to use the audit data.

Result list displaying past activities based on filter criteria
Result list displaying past activities based on filter criteria

The interface groups objects into three execution states:

  • Succeeded
  • Failed
  • Cancelled

The basic filter options permit filtering on predefined time periods or a user-defined custom period. As with the current activities task, the result set can be narrowed down to objects run by an individual user.

Upcoming activities task

This task has gone through changes in the IBM Cognos release. An interactive chart was introduced to better represent the estimated pending load on the system. This is accomplished by displaying an hourly breakdown of the amount of objects that have been scheduled to start in that particular hour timeslot.

Hourly breakdown of scheduled activities in the system
Hourly breakdown of scheduled activities in the system

In addition to the filter options on the left part of the interface, it is possible to scroll between days by using the Next and Previous buttons beside the date at the top of the chart. Beneath the chart is a filtered list of the individual activities that are scheduled to execute. Clicking a specific hour bar in the chart automatically filters the results in the list. To provide visual context to the filtered list, the currently applied filters are included above the list .

List of scheduled objects based on the criteria set in the chart
List of scheduled objects based on the criteria set in the chart

When creating a schedule, privileged users can assign a priority of 1 (high) to 5 (low) for the object. All other users have a default priority of 3 when defining a schedule. Priority indicates in which order the dispatcher processes requests when items are queued. That is, in a busy environment where report executions are submitted when the engine is at capacity, they will be placed into a queue and processed when the engine resources become available. When the engine is ready to handle a new request it takes the oldest priority 1 request before it takes requests with a priority of 2 or higher.

Priority reports: Priority does not mean that a long-running priority 5 report execution, in the process of executing, is terminated when a priority 1 request is received. It means that the priority 1 request is executed first after the priority 5 request has completed executing.

System task

This task provides administrators with an overall glimpse of how the system is faring through the use of status indicators on a dashboard. The System task interface is divided into three fragments:

  • Scorecard: Left frame that displays a summary of the overall health of the components that make up the environment.
  • Metrics: Upper-right frame that lists all of the metrics, and their score, that pertains to the object in focus (from the scorecard fragment). The default is to display the metrics for the overall system (environment).
  • Settings: Lower-right frame that shows a read-only view of the configuration parameters that pertain to the object in focus.

The scorecard frame provides two views:

  • The default standard view, which allows administrators to navigate through the system topology to verify the health of the servers, dispatchers, and services, as well as their corresponding metrics
  • The comparative view, which provides the ability to watch predefined key metrics as they relate to similar objects.
  • Scorecard view of the servers in the topology
    Scorecard view of the servers in the topology

Figure provides administrators with a view of the entire IBM Cognos BI topology. The sample shows that there are two servers that make up this particular environment. Besides the indicator lights that show the overall health of the object based on tolerance thresholds that can be applied to the metrics, the status of the object is also displayed. Industry standard terms employed are:

  • Partially available: Indicates that one or more of its children is unavailable
  • Available:Indicates that the object is online
  • Unavailable:Indicates that the object is offline or is not started

In the previous example, the wottmassons server is online and available, while the ca093489 server is partially available. The net effect is that the overall system status becomes partially available due to the fact that at least one child is unavailable. The underlying dispatchers and services can be viewed by drilling down on the server name, which maintains the parent/child relationships. For example, drilling down on the ca093489 server name reveals the dispatchers running on that server.

Server dispatchers, each with a different status
Server dispatchers, each with a different status

Besides the ability to drill up and down, using the Change view button beside an entry allows the results to be filtered to obtain the desired view. This drop-down menu allows for filtering of all servers, server groups, dispatchers, and services, as well as selecting individual services. Selecting a service, for example, report service, provides a list of all report services, their overall health, and their current state. The ability to perform metric comparisons is not be possible though unless you switch focus back and forth between report services.

Monitoring services side by side in the Scorecard view does not provide the ability to easily compare metrics across similar services in the environment. There is, however, a comparative view that allows administrators to see certain predefined metrics as they pertain to related objects in one single view.

To access this view, expand the scorecard frame by clicking Maximize. This maximizes the frame within the interface and provides a comparative view.

Comparative view of services in the environment
Comparative view of services in the environment

At first glance it is difficult to determine which dispatcher each of the report services belongs to. The parent dispatcher is visible by hovering over the service icon, which produces a tooltip with the parent’s name.

Tool tip on icon displays the dispatcher name
Tool tip on icon displays the dispatcher name

Schedules task

The schedules task displays all of the active (enabled or disabled) schedules in the environment and which user scheduled the object. The summary chart is grouped by status.

Summary chart displaying the status of the schedules
Summary chart displaying the status of the schedules

The result list shows the scheduled objects along with the last modified date, who created the schedule, and the status and the priority assigned to the schedule.

Detailed list of schedules in the system
Detailed list of schedules in the system

Security tab

The Security tab contains all of the tasks that are required for managing the namespaces and capabilities.

Users, groups, and roles task

This task allows administrators to create and manage application-specific groups and roles in the built-in Cognos namespace. In addition, the external third-party security namespaces can be browsed and the contents of the users’ My Folders area can be viewed.

Capabilities task

Capabilities, not to be confused with security, is a collection of features that can be granted, or denied, to the user community. The capabilities task is the tool that allows administrators to set the global capabilities that dictate which features and functionality are available to the users logging in to access the application. The IBM Cognos BI release allows for a customizable approach to administration and responsibility so that various areas of responsibility can be delegated to more focused administrators. This helps to effectively distribute the administrative duties across users and groups.

Configuration tab

TheConfiguration tab is a collection of administrative tasks pertaining to managing the content and the various aspects of the environment. Items such as managing printers, data source connections, and configuring parameters as they relate to the dispatchers and services that make up the IBM Cognos topology, are located on this tab.

Data source connections task

The data source connections task permits administrators to define and manage connections to data sources and the sign on credentials associated with them.

Content administration task

This task is the interface that allows for the creation and management of content import and export definitions. In addition to the content deployment objects, this task also provides the ability to create consistency checks and enhanced search index updates, and then either execute or schedule them.

Printers task

This task is where printer connections are defined, created, and managed for the IBM Cognos BI solution.

Styles task

This task allows administrators responsible for the IBM Cognos Connection portal to assign privileges to the various styles (skins) available to the user community. Access can be granted or denied to each style in the list as required. Denying access to a particular style for a user removes the ability to select that style as part of the user’s preferences.

Portlets task

The portlets task is the mechanism within the product that allows administrators to manage and control the access rights to the portlets that are part of the IBM Cognos BI solution.

Dispatchers and services task

This task provides server administrators with a tool to manage the configuration settings of the dispatchers and services in the IBM Cognos topology. Configuration parameters can be set at the highest level and pushed down to the individual dispatchers through acquired properties, or they can be set individually on each dispatcher independent of the parent settings.

Index Search tab

TheIndex Search tab provides administrators with the settings and parameters required to create and manage the enhanced search index.This index provides search results to enhanced consumers when executing searches to answer key business questions.

All rights reserved © 2018 Wisdom IT Services India Pvt. Ltd Protection Status

IBM Cognos Topics