Monitoring data grids in the user interface

You can use the charting capabilities of WebSphere® DataPower® XC10 Appliance to view the overall performance of the data grids in your environment.

Before you begin

After you create your data grids and configure your applications to use the data grids, allow some time to pass for the statistics to become available. For example, with a dynamic cache data grid, statistics are not available until a WebSphere Application Server that is running a dynamic cache connects to the dynamic cache data grid on the appliance. If you are using a collective, the collective initialization must be complete before statistics are available. In general, wait up to one minute after a major configuration change to see the changes in your statistics.

About this task

Charting behavior
The day, week and month data series are persisted long-term. However, the Last hour data series is only stored in memory. The last hour data is stored in the following locations:
  • In memory for a data grid
  • In memory for the console cache

If a piece of statistics data is lost in only one of these memory locations, then the data is still available. However, if the data is lost in both locations, the data does not display in the chart.

Memory data loss conditions:
  • When data grids are moved, time periods occur when statistics are not accumulated.
  • When appliances are restarted, the memory data for both the data grid and the console cache is cleared.
  • When the memory for the console cache fills, the least recently used data is evicted from the console cache. The console cache holds 2048 statistics entries.
  • When the console is inactive for 30 minutes, the console process restarts and the memory is cleared.

The collection of statistics will continues to re-accumulate for the last hour data series.

Tip: To view more specific information about any data point in a chart, you can move the mouse pointer over the data point.

Procedure

  • To view the performance of all of your data grids, click Monitor > Data grid overview. This page contains the following information:
    Used capacity tab
    Current data grid used capacity distribution
    This chart contains a view of the total amount of capacity that is available on the appliance or collective, the total amount of capacity that is being used by primary and replica copies of data, the remaining limited capacity if a capacity limit has been set, and the data grids that are consuming the most capacity. You can use the following options to sort the data. Only the top 25 data grids are displayed:
    • Largest used capacity consumers
    • Largest percent of limited capacity used
    Used capacity over time
    This chart displays the used capacity over the selected time period.
    Average throughput tab
    Top 5 data grids most active data grids by average throughput in transactions/second
    This chart contains a list of the top 5 data grids, organized by the average throughput, which is measured in transactions per second.
    Average throughput over time
    This chart displays the average throughput, which is measured in transactions per second, over the selected time period.
    Average transaction time tab
    5 slowest data grids by average transaction time in milliseconds
    This chart contains a list of the slowest five data grids, organized by the average transaction time.
    Average transaction time over time
    This chart displays the average transaction time over the selected time period.
  • To view individual data grids, click Monitor > Individual data grid overview > data_grid_name. This page shows a summary that includes the number of cache entries, the average transaction time, average throughput, cache hit rate, and the percentage of limited capacity over the last 30 seconds. You can also view the following charts:
    Used Capacity
    This chart shows the used capacity of the cache versus the actual number of entries and the configured capacity limit of the cache. You can edit the time range that is displayed: last hour, last day, last week, last month. The level of detail that is showed on the chart varies depending on the time range that you select.
    Interpreting the number of cache entries:
    • For dynamic cache data grids: 94 cache entries are created by default for every dynamic cache data grid. These cache entries contain dynamic cache provider statistics and the dynamic cache configuration for WebSphere Application Server. Therefore, 94 cache entries are displayed in the monitoring panel before you add any data to the dynamic cache data grid.
    • [Version 2.0.0.3 and later] For dynamic cache data grids: 166 cache entries are created by default for every dynamic cache data grid. Each dynamic cache grid has 83 partitions, and each partition or shard is initialized with two entries for dynamic cache grids. Therefore, the number of cache entries at initialization is 166. These cache entries contain dynamic cache provider statistics and the dynamic cache configuration for WebSphere Application Server. Therefore, 166 cache entries are displayed in the monitoring panel before you add any data to the dynamic cache data grid.
    • For session data grids: The number of entries in the data grid includes counts for the number of sessions, the number of attributes across all sessions, and eviction table entries. The eviction table is populated when a timed out session that has not been invalidated by the Web container exists. The eviction table is only populated during an application server failover.
    Cache Usage
    This chart helps to visualize the number of successful queries to the cache. You can display cache attempts, cache hits, and the cache hit rate in the chart.
    Average Throughput
    This chart displays the average number of transactions per second that are being processed over a given time range and the average length of time for each transaction.
  • To view further details about a specific data grid, click Monitor > Data grid detail reports. A tree displays with all of the data grids in your configuration. You can drill down into a specific data grid to view the maps that are a part of that data grid. You can either click a data grid name or a map for further information:
    Data grid details
    You can view the used capacity, and a list of zones to which the data grid belongs. A chart that shows the capacity consumed by the top 25 maps in the data grid displays. You can also view a total pool, which includes the capacity by zone. A chart shows how much data grid capacity is consumed across the top 25 zones.
    Map details
    You can view further information about the maps in each grid, including the following details: total number of cache entries in the map, average throughput, average transaction time, and the total capacity of the map broken down by the top 25 partitions.
Parent topic: Monitoring