Creating caches with Data Grid Operator

  • Red Hat Data Grid 8.3
  • Updated 01 December 2022
  • Published 06 December 2021

Creating caches with Data Grid Operator

Red Hat Data Grid 8.3
  • Updated 01 December 2022
  • Published 06 December 2021

Use Cache CRs to add cache configuration with Data Grid Operator and control how Data Grid stores your data.

Data Grid caches

Cache configuration defines the characteristics and features of the data store and must be valid with the Data Grid schema. Data Grid recommends creating standalone files in XML or JSON format that define your cache configuration. You should separate Data Grid configuration from application code for easier validation and to avoid the situation where you need to maintain XML snippets in Java or some other client language.

To create caches with Data Grid clusters running on OpenShift, you should:

  • Use Cache CR as the mechanism for creating caches through the OpenShift front end.

  • Use Batch CR to create multiple caches at a time from standalone configuration files.

  • Access Data Grid Console and create caches in XML or JSON format.

You can use Hot Rod or HTTP clients but Data Grid recommends Cache CR or Batch CR unless your specific use case requires programmatic remote cache creation.

Cache CRs

  • Cache CRs apply to Data Grid service pods only.

  • Each Cache CR corresponds to a single cache on the Data Grid cluster.

Creating caches with the Cache CR

Complete the following steps to create caches on Data Grid service clusters using valid configuration in XML or YAML format.

  1. Create a Cache CR with a unique value in the field.

  2. Specify the target Data Grid cluster with the spec.clusterName field.

  3. Name your cache with the field.

    The name attribute in the cache configuration does not take effect. If you do not specify a name with the field then the cache uses the value of the field.

  4. Add a cache configuration with the spec.template field.

  5. Apply the Cache CR, for example:

    oc apply -f mycache.yaml created

Cache CR examples

kind: Cache
  name: mycachedefinition
  clusterName: infinispan
  name: myXMLcache
  template: <distributed-cache mode="SYNC" statistics="true"><encoding media-type="application/x-protostream"/><persistence><file-store/></persistence></distributed-cache>
kind: Cache
  name: mycachedefinition
  clusterName: infinispan
  name: myYAMLcache
  template: |-
      mode: "SYNC"
      owners: "2"
      statistics: "true"
        mediaType: "application/x-protostream"
        fileStore: ~

Adding persistent cache stores

You can add persistent cache stores to Data Grid service pods to save data to the persistent volume.

Data Grid creates a Single File cache store, .dat file, in the /opt/infinispan/server/data directory.

  • Add the <file-store/> element to the persistence configuration in your Data Grid cache, as in the following example:

    <distributed-cache name="persistent-cache" mode="SYNC">
      <encoding media-type="application/x-protostream"/>

Adding caches to Cache service pods

Cache service pods include a default cache configuration with recommended settings. This default cache lets you start using Data Grid without the need to create caches.

Because the default cache provides recommended settings, you should create caches only as copies of the default. If you want multiple custom caches you should create Data Grid service pods instead of Cache service pods.

  • Access the Data Grid Console and provide a copy of the default configuration in XML or JSON format.

  • Use the Data Grid CLI to create a copy from the default cache as follows:

    [//containers/default]> create cache --template=default mycache

Default cache configuration

This topic describes default cache configuration for Cache service pods.

<distributed-cache name="default"
  <memory storage="OFF_HEAP"
          when-full="REMOVE" />
  <partition-handling when-split="ALLOW_READ_WRITES"

Default caches:

  • Use synchronous distribution to store data across the cluster.

  • Create two replicas of each entry on the cluster.

  • Store cache entries as bytes in native memory (off-heap).

  • Define the maximum size for the data container in bytes. Data Grid Operator calculates the maximum size when it creates pods.

  • Evict cache entries to control the size of the data container. You can enable automatic scaling so that Data Grid Operator adds pods when memory usage increases instead of removing entries.

  • Use a conflict resolution strategy that allows read and write operations for cache entries, even if segment owners are in different partitions.

  • Specify a merge policy that removes entries from the cache when Data Grid detects conflicts.