Data caching

Principle of work

Between the basic data repository dataRepo and the data repository with security checks securedDataRepo, a data repository with support for caching cachedDataRepo has been implemented. The data is loaded by chain: BD -> dataRepo -> cachedDataRepo -> securedDataRepo.

However, when running a query, the cachedDataRepofirst checks for the availability of data in the cache (if the cache is enabled), and if there is no data in the cache, it requests it from dataRepo, then it places it in the cache and returns it to securedDataRepo.

Setting caching of objects of individual classes

In the configuration file (deploy.json), in the options cachedDataRepo you can individually specify the list of classes whose data should be stored in the cache.

Example:

...
"options": {
  "cachedClasses": ["class1@namespace", "class2@namespace", "class3@namespace"]
}
...

If cachedClasses is not specified, then all data are cached.

Caching a list of objects

Both individual objects and lists are cached.

When a list is cached based on its selection conditions (including pagination), a key is generated, then a selection is performed and all received objects are cached individually, and the list is cached as an array of object identifiers. When selecting a list from the cache, on the basis of this array, the corresponding objects are selected from the cache and the resulting list is formed.

Similarly, all objects by reference and eagerly loaded collections are cached.

Setting the adaptive cache

For advanced configuration of cache (such as, query storage time, depth of storage of objects in the cache, etc.) the setting of adaptive object caching is applied. Adaptive cache settings are specified in the deploy.json file in the "di"property, in the memcached and redis components in the connectOptions option. All the necessary placeholders are listed in the config.json file of the platform repository.

NB: The storage depth of objects in the cache corresponds to the depth of the query for objects from the database. So, the object stores information about its references, and then in the cache the object graph is obtained, and not the tree.

Setting the cache by use of the ini-file

The ini-file parameters:

...
cache.module=redis // - модуль используемый репозиторием данных для кеширования.
//(варианты: ion://memcached, ion://redis, ion://innerCache)

cache.memcached.enabled=true // активируем memcached
cache.memcached.location1 // сервер1
cache.memcached.location2 // сервер2
cache.memcached.timeout // таймаут обращения к кэшу

cache.redis.enabled=true // активируем redis
cache.redis.host=127.0.0.1 // хост redis
cache.redis.port=6379 // порт redis

session.type=redis // хранение сессий авторизации в redis

Cache is not used by default.

---