Application - Top Applications
In order to make sure that a Citrix/Terminal server performs at its peak at all occasions, it is essential to keep tabs on the number of applications executing on the server and their combined and individual resource usage. If the most popular applications in a farm can be identified, administrators can consider alternative load balancing strategies to ensure that the most popular or most resource intensive applications are hosted on different servers.
The Top Applications Report provided by eG Enterprise facilitates a comprehensive review of the usage of applications hosted on a Citrix/Terminal server farm. By allowing administrators to configure the application-specific metrics that they want displayed in the report, the eG Enterprise suite ensures that useful and workable information is at the disposal of the administrators at all times. As the administrators get to see what they want to see, they will be able to take more effective performance decisions.
In order to generate the application report, do the following:
Follow the menu sequence: REPORTS BY FUNCTION -> Domain Specific Reports -> Citrix XenApp/XenDesktop -> Applications -> Top Applications.
The Application - Top Applications page will then appear. To monitor the servers in a farm, select the Zone to which the servers in the farm have been added.
Choose a component type for which the application activity is to be monitored from the Component Type list box.
If Citrix XenApp is chosen from the Component Type list, then an additional Worker Group list box will appear. Worker groups are collections of XenApp servers, residing in the same farm, that are managed as a single unit. You can publish applications to a worker group. Select a worker group from this list if you wish to view the usage of a User belonging to that particular worker group. By default, Any option is selected from this list.
Note:
The Worker Group list box will appear only if the SHOW WORKER GROUPS flag is set to Yes while configuring the Citrix Applications test.
Next, choose a Report Type (Graph or Data). The default Report Type is Graph. To closely observe and clearly understand the behavior of a single performance parameter across a zone, you can plot a single measure's data alone in a Graph report.
If the Report Type is Graph, a Measure list box will appear, containing the complete list of measures that you have configured for the report. From this list, pick the Measure for which a graph is to be generated.
Then, specify the Timeline for the graph. You can either provide a fixed time line such as 1 hour, 2 days, etc., or select the Any option from the list to provide a From and To date/time for report generation.
Note:
The default Timeline can be set using the CONSOLIDATED REPORTS page in the eG administrative interface. This page appears when you follow the Admin tab -> Settings -> Reporter -> Consolidated Reports menu in the eG administrative interface.
Shorter the Timeline for a report, lighter would be the query that is executed on the database, and faster the report generation. Report Timelines that run across many weeks therefore, cause excessive strain on the database, thus slowing-down report generation. To save the eG backend from injury, eG Enterprise automatically restricts the maximum Timeline for which a thin client report can be generated to the lower of the following values:
the value of the CtxLoginTest parameter in the [DDD_CLEANUP] section of the eg_format.ini file (in the <EG_INSTALL_DIR>\manager\config directory)
the value of the MaxTimeline parameter in the [MISC] section of the eg_report.ini file (in the <EG_INSTALL_DIR>\manager\config directory).
The maximum timeline for which the current user to eG Reporter is permitted to generate reports;
The [DDD_CLEANUP] section is where you define the interval (in days) at which the detailed diagnosis of specific tests is to be purged from the database. For the CtxLoginTest, the default interval is set to 32 days (i.e., just over a month), indicating that, by default, the detailed diagnosis of the CtxLoginTest is cleared from the database at the end of every month. Since critical thin client reports are generated based on the detailed diagnosis information, a Timeline that exceeds the duration set against CtxLoginTest would result in a report that displays invalid/no data. This duration however, is configurable. Similarly, you can also configure a MaxTimeline in the [MISC] section of the eg_report.ini file. By default, this parameter is set to 168 (hours), indicating that the maximum Timeline for a thin client report is 1 week (i.e., 24 * 7 = 168 hours). Also, while creating user profiles using the eG administrative interface, an administrator can indicate the maximum timeline for which that user can generate reports, so as to avoid long running reports.
Now, assume that you have set the CtxLoginTest parameter to 21 days (i.e., 3 weeks) and the MaxTimeline parameter to 336 hours (i.e., 2 weeks). Also, say that user john with access rights to eG Reporter has been configured to generate reports for a maximum timeline of 1 week. Therefore, whenever user john logs into eG Reporter and attempts to generate a thin client report, eG Enterprise compares the three values, picks the lowest of the three, and sets that value as the maximum Timeline for thin client reports. In our example, the maximum timeline settings of user john (i.e., 1 week) is the lowest Therefore, the maximum value that will be available for selection from the Fixed Timeline list box in the thin client report page will be 1 week.
Note that this timeline restriction does not apply to fixed timeline settings alone. If a user chooses the Any option from the Timeline list and proceeds to provide a start date and end date for report generation using the From and To specifications, eG Enterprise will first check if the user's Timeline specification is equal to or below the lowest of the three values discussed above. If not, report generation will fail. For instance, for a user who is allowed to generate reports spanning over a maximum period of 3 days only, the difference between the From and To dates should never be over 3 days. If it is, then, upon clicking the Run Report button a message box will appear, prompting the user to change the From and To specification.
In addition to the settings discussed above, this report comes with a set of default specifications. These settings are hidden by default. If you do not want to disturb these default settings, then you can proceed to generate the report by clicking the Run Report button soon after you pick a Measure. However, if you want to view and then alter these settings (if required), click on the
button. The default settings will then appear in the More Options pop up window. The steps below discuss each of these settings and how they can be customized.Specify the start time and end time for report generation against the Time period field.
Note:
The default Time period to be displayed here can be configured using the CONSOLIDATED REPORTS page in the eG administrative interface. This page appears when you follow the Admin tab -> Settings -> Reporter -> Consolidated Reports menu in the eG administrative interface.
If the timeline specified for the report needs to exclude the data collected during the Weekends, then set Exclude weekends to Yes. If not, select No.
Note:
You can configure the days of the week that need to be considered as a ‘weekend’ using the CONSOLIDATED REPORTS page in the eG administrative interface. This page appears when you follow the Admin tab -> Settings -> Reporter -> Consolidated Reports menu in the eG administrative interface.
Using the Weighted Average list box, you can indicate how the Avg value is to be computed for a chosen measure for the purpose of this report. The status of the Weighted Average flag is relevant only if the Test chosen is a descriptor-based test, and the descriptors are dynamic in nature. For instance, the Citrix Applications test in our example auto-discovers the applications that are currently in use on a Citrix XenApp farm. For each application that is available in the Citrix XenApp, the test reports a variety of statistics. Unlike descriptors such as disk partitions or processors that rarely change, the applications are dynamic descriptors, which may change often; in other words, an application that is currently in use by a user, may not be used at all during the next measurement period. The Citrix Applications test will neither report metrics for the inactive descriptors nor display it in the eG monitoring console. This is why, applications in the Citrix Applications test are considered ‘dynamic descriptors‘.
By default, this report compares the Avg value of the chosen measure across all selected descriptors. Since the Weighted Average flag is set to No by default, this Avg is computed as the ratio of the sum total of the measure values reported by a descriptor during the given timeline and the total number of times the test is executed during the same timeline. In case of dynamic descriptors however, the Avg values so computed may not reveal the ‘true picture of performance’. This is because, the test may not discover or report metrics for dynamic descriptors throughout a given timeline. For example, take the case of the Citrix Applications test. Say, two applications - namely, ‘A’ and ‘B’ are currently in use in the Citrix XenApp server. During its first measurement period; the CPU Utilization of an application is registered as 2% and 8% respectively. Assume that the second time the Citrix Applications test ran, it captured 5% as the CPU Utilization of application ‘A’. Application ‘B’ however was not used during the second measurement period, and hence, was not discovered at all. If the Weighted Average is set to No by default, then, this report will plot the Avg value of 4% (3+5=8/2=4) for application ‘A’ and 4% (8/2=4) again for application ‘B’. If you notice, unlike application ‘A’, where 8% of CPU was utilized over a period of time, in case of application ‘B’, 8% of CPU was utilized at one shot! Logically therefore, application ‘B’ has to be ranked above application ‘A’ in terms of CPU utilization. However, since the default Avg value computation does not clearly bring out this difference, both application ‘A’ and ‘B’ are treated at par in this report! This is why, in case of dynamic descriptors, you may want to set the Weighted Average flag to Yes. In this case, the eG Enterprise system expresses Avg as the ratio of the sum total of the measure values reported by a descriptor during a given timeline and the ‘total number of times that descriptor was active’ during the same timeline. This implies that if the Weighted Average is set to Yes in the example above, the Avg value for application ‘A’ will continue to be 4 (3+5=8/2=4), but the same for application ‘B’ will be 8 MB (8/1=8). In this report therefore, application ‘B’ will be placed above application ‘A’, thereby accurately pointing you to the top application that is consuming maximum CPU resources.
Select the application that needs to be excluded during report generation using Exclude Applications list box.
In large environments, while generating reports for long time periods, say a few weeks to months, report generation was found to slow down, owing to the large volume of data being processed. If users needed to generate many such reports quickly, the slowdown compelled them to generate these reports one after another, thus affecting the speed of their operations, and consequently, their productivity. eG Enterprise now includes a background processing feature that will enable users to process multiple reports, spanning long time periods in the background, while allowing them the bandwidth to generate short-term reports in the foreground. This way, users can generate multiple reports with little-to-no loss of time. To schedule background processing of a report, select the Background Save - PDF option from the Report Generation list. In this case, a Report Name text box will appear, where you would have to provide the name with which the report is to be saved in the background. To process reports in the foreground, select the Foreground Generation - HTML option from this list.
Note:
The Report Generation list will appear only if the EnableBackgroundReport flag in the [BACKGROUND_PROCESS] section of the eg_report.ini file (in the {EG_INSTALL_DIR}\manager\config directory) is set to Yes.
The default selection in the Report Generation list will change according to the Timeline specified for the report. If the Timeline set is greater than or equal to the number of days specified against the MinDurationForReport parameter in the [BACKGROUND_PROCESS] section of the eg_report.ini file, then the default selection in the Report Generation list will be Background. On the other hand, if the Timeline set for the report is lesser than the value of the MinDurationForReport parameter, then the default selection in the Report Generation list will be Foreground. This is because, the MinDurationForReport setting governs, when reports are to be processed in the background. By default, this parameter is set to 2 weeks - this indicates that by default, reports with a timeline of 2 weeks and above will be processed in the background.
Finally, click the Run Report button to generate the report.
If the Report Type is Graph and the option chosen from the Report Generation list is Foreground Generation - HTML, then, clicking on the Run Report button will invoke the Thin client Application Activity report revealing the following:
A distribution pie chart revealing the number of components that fall within the various value ranges arrived at for the selected Measure; the values are computed by applying the first of the configured functions on the chosen Measure. For instance, assume that Memory utilization is the chosen Measure. Say that you have configured to display the Avg and Max of Memory utilization in the application report. Typically, both these configured values will appear only in a Data report. In the case of a Graph report however, the first of the two functions - i.e., Avg of Memory utilization - is alone calculated for the chosen zone. The resulting values are then grouped as value ranges, and the number of components that fall within each of the value ranges are identified and displayed in a pie chart. A quick glance at the pie chart indicates where memory has been excessively spent. Click on a particular slice in the pie chart to view the list of components that fall within the value range represented by that slice. Against every application name, the actual values for each of the configured functions (both Avg and Max, in our example) will be displayed.
Note:
By default, the chart type for distribution is a pie chart. However, you can have a bar graph depict the same data instead of a pie chart, by following the steps given below:
Edit the eg_report.ini file in the {EG_INSTALL_DIR}\manager\config directory.
Change the value of the chartTypeForDist parameter in the [APPLICATION_REPORT] section of this file from the default Pie, to Bar.
Save the eg_report.ini file.
By default, the number of value ranges that need to be configured for the distribution chart is 10. To override this default setting:
Edit the eg_report.ini file in the <EG_INSTALL_DIR>\manager\config directory.
Specify a number of your choice against the nofRangeForDist parameter in the [APPLICATION_REPORT] section of this file. By default, this parameter will hold the value 10.
Save the eg_report.ini file.
Below the pie chart, you will find a bar chart that indicates the applications that have topped/failed in a selected performance realm (i.e., the Measure) during the specified Timeline. For example, for the Memory usage measure, this bar chart reveals the top memory-consuming applications in the zone. Like the pie chart, the values for the bar chart are also calculated by applying the first of the configured functions on the chosen Measure.
Note:
The number of components to be displayed in the Top <N> Components bar chart is configurable. To specify the number, do the following:
Edit the eg_report.ini file in the <EG_INSTALL_DIR>\manager\config directory.
- Specify a number of your choice against the NoOfServersforChart parameter in the [APPLICATION_REPORT] section of the file. By default, the value displayed here is 10, indicating that, by default, the bar chart will be for the Top 10 Components. If you change this value to 5, then a bar chart displaying the Top 5 Components will appear.
Finally, save the eg_report.ini file.
Note:
You can configure the colors to be used in the distribution chart and the Top <N> Components bar chart in the APPLICATION report, by editing the eg_report.ini file in the <EG_INSTALL_DIR>\manager\config directory. The [APPLICATION_REPORT] section of the file defines the 20 default colors of the distribution and bar chartsof an APPLICATION report, in the following format:
[APPLICATION_REPORT]
ChartColor=#E3A869, #FFCCCC, #A08583, #CAAFCA, #B0ABB2, #9FB6CD, #A78AAD, #FFB2C0, #F99B90, #CC9999, #D8C7CD, #D2B48C, #F4DF89, #E291BB, #FF9966, #CCCC99, #B5B072, #CCFFFF, #F4E9BC, #CC9966
distColor=#FFDAB9, #B5A642, #48D1CC, #E9967A, #9BCD9B, #DEB887, #CC99B3, #F2D49C, #A9A9A9, #CCCC99, #EED3D3, #99CCCC, #E5FFF2, #99B3CC, #E7E7F9, #CCB399, #B3DAFF, #E4AF95, #D3E7C1, #FFB3B4
You can override the default color settings of the distribution chart by modifying the color-codes specified against distColor. For changing the colors used by the Top <N> Components chart, alter the codes listed against the ChartColor parameter.
A Data report on the other hand, does not graphically represent the measure data. Instead, the configured details are presented in a tabular format in the report.
If you choose Data as the Report Type, then an additional Component list box will appear allowing you to select the components of your interest. By default, all components of the chosen Component Type will be selected and displayed in the Component list box. To monitor the applications executing on all servers in the chosen Zone, select ALL option from the Component Type list box. Doing so, all the servers in the chosen Zone will be selected and listed in the Component list box.
Next, choose a specific Application whose activity is to be monitored.
To sort the resulting Data report by a particular column, choose the column name from the Sort by list.
Then, like in the case of a Graph report, provide a Time period, Timeline, and indicate whether/not to Exclude weekends. Also, pick the Foreground Generation - HTML option from the Report Generation list. Finally, click on the Run Report button.
In case of Data report, a table providing the critical performance statistics for the chosen server will appear. By default, the following metrics will be displayed:
- The Avg number of instances of an application executing on the chosen server
- The Avg CPU utilization of every application
- The Avg memory utilization of every application
While these default metrics cannot be changed, you can, however, configure functions other than Avg to be executed on the values reported by these metrics. Similarly, additional metrics can also be configured for display in the report.
Besides the default and configured measure information, the Data report displays the percentage of time for which the chosen application(s) was in use, and the number of distinct (unique) users who were accessing the application(s) during the specified timeline. This information clearly indicates the least and the most used applications on the server.
Also, to view the top-n users of the application at the server-level, click on the Top <N> users link corresponding to the application. The top-n users of the application on the chosen Server will then be displayed.
If, at step 17 above, any of the ‘All . . .’ options had been chosen (either ALL option is chosen from the Component Type list or all the individual servers are selected from the Component list) instead of a particular Component Type and Component, then an additional Aggregate across servers field will appear.
While monitoring the applications executing on all the servers in a chosen Zone, you can indicate whether the measures to be displayed in the report are to be aggregated at the zone-level or the individual server-level. Selecting the Yes option against Aggregate across servers will ensure that the usage metrics displayed in the report for the chosen application(s) is aggregated across all the servers in the selected zone. Clicking on the Top <N> users link corresponding to an application in this report will therefore reveal the top-n users of the application across the entire zone.
On the other hand, if you select No against the Aggregate across servers flag, then the report will reflect the usage of every chosen application(s) on every server in the selected zone - in other words, the report will display a different set of metrics for every server-application combination. In this case therefore, clicking on the Top <N> users link that corresponds to a server-application combination will list the top-n users of that application on that server.
If the Background Save - PDF option is chosen from the Report Generation list, then clicking on the Run Report button will not generate the report and display it in this page for your benefit. Instead, a message indicating that the report is being processed in the background will appear. This will be accompanied by a link, that will lead you to the page that lists all the reports that are being processed in the background, and their current status. If background report generation fails for a report, you can regenerate that report using this page, or can even delete that report if need be. On the other hand, if background processing successfully completes for your report, then, you can view a PDF of the report by clicking on the PDF icon in that page.
- Using the icons available at the right, top corner of the report, you can do the following:
- PRINT the report
- SAVE the report as a PDF file
- MAIL the report to configured individuals
- Save a Data report to a CSV file
- Add the report as a FAVORITE
- SCHEDULE the automatic mailing of the report at configured intervals to specific recipients