Follow

Statistics Portlet

The Statistics Portlet of the Heirloom Dashboard provides statistics on cloud-based applications you currently have running. Access the Heirloom Dashboard through paas.heirloomcomputing.com > Sign In > Dashboard.  This portlet provides information for you and your group (if you've been designated a group administrator of your group)

dashboard-statistics-1.png

Statistics are requested by righ-clicking on one or more instances shown in the Instance Portlet.  Statistics can be displayed on a number of metrics, units, period over which time the data is collected, date ranges and instances to which they pertain.  Typically data are collected over a period of 5 to 15 minutes for small duration (hour, daily) periods of time and hourly for charts that show an entire week's worth of data.  Bring up a specific value by hovering over a datapoint within most charts. 

Averages are depicted with a line chart, as shown above.  When a range is requested, a column chart shows the upper (maximum) range, lower (minimum) range with a dot indicating the mean (average) within that range. 

dashboard-statistics-2.png

 

The following statistics are available:

  • Show CPU Utilization - the current and past CPU load (in CPU percentage) for the instance.
  • Show CPU Utilization (range) - the minimum, maximum and average CPU load.
  • Show Network Usage - the current and past network output (in total bytes) being handled by the instance.
  • Show Network Usage (range) - the minimum, maximum and average Network Usage.
  • Show Response Time - for secure (https) applications with statistics tracking enabled, display the average response time seen by end-users of your application.
  • Show Response Time (range) - the minimum, maximum and average Response Time within the interval
  • Show Requests (range) - the minimum, maximum and average Web requests (by number and data size) issued by your users within the interval
  • Show Responses (range) - the minimum, maximum and average Web pages (by number and data size) generated to your users within the interval.

The bar chart's interval depends on the duration of the graph selected by the refresh "gear" tool in the statistics portlet.  An hourly duration, for example, 13 bars are shown and each bar represents statistics within a 5 minute period on 5-minute boundaries (:00, :05, etc.).  Two partial 5-minute periods for the last hour are shown, resulting in 13 bars. Statistics that are gathered at precisely 14:23:35 in the afternoon represent statistics for 13:20:00 through 14:24:59, the last one being partial.  For longer duration graphs (daily, weekly) the interval is 15-minutes or hours.

Network Usage statistics shows amount of data delivered via Web access to end users, access to remote databases (include Hybrid Clouds to your on-premise datacenter) and any messaging traffic between cooperating instances in a CICS cluster.

dashboard-statistics-3.png

Response Time measurements and specific Request and Response statistics are available when accessing secure Web content (https), the preferred approach in instances that have been configured for such access.  The number of requests/response as well as the range or average data packet size per request/response is provided.

 

Click on the links in the upper right hand corner of the portlet to change or refresh the view.  
  • icon-minimize.png minimize the view.  You can drag and drop the view to reposition it on the dashboard
  • icon-refresh.png refresh the statistics in four different time spans (4 buttons).  Select statistics over an hour, 3 hours, over the 24 hour period and one week
  • icon-select.png select (2 buttons) either statistics specific to your ELPaaS or Elastic COBOL login ID or your groups
  • icon-close.png close the portlet.  You can reopen by right-clicking anywhere on the page and choosing Summary Portlet from the menu

Programmer Note:  Detailed statistics with advanced ranges are available for instances programmatically through the Heirloom Computing REST API.

Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request

1 Comments

  • 0
    Avatar
    Mark

    Think of the bar in a range chart as showing the upper and lower limits within that interval and the red line the average.  A red line in the middle of the bar means the system is well balanced.  A red line nearer the lower range indicates there are many peaks within the interval -- extending the maximum -- but the greater number of samples is closer to or below the average.  Note that the median is not shown ... the point at which there are an equal number of samples below and above the line ... but in practice the median is close to the average on most of these statistics.

Please sign in to leave a comment.
Powered by Zendesk