Group Cache Data: An Overview

Note

This feature is only available in NCache Enterprise Edition.

Items in a cache having a relation between them or falling under the same category can be grouped in NCache using NCache "Groups" feature. Using Groups, you can logically partition your data for better efficiency.

When to Use Groups?

Let's assume a scenario where you have a significant amount of customers in your company, and you want to categorize the customers according to the revenue generated. For instance, the first category should add all those customers whose generated revenue is above $1000 as Important Customers to the cache.

This can be done by executing an SQL query over the database that obtains the result set containing those customers. However, to save the effort of executing the SQL query every time on the entire database, you can add the result set to the cache with a group name once, so the data resides under that particular group.

The table below is an illustration of the above scenario.

using Groups in NCache

Why Use Groups?

Groups improve efficiency and provide the user with ease of fetching or removing data based on logical categories. When the cached data is grouped according to certain rational criteria, the time to search the data falling under the same criteria significantly reduces. Hence, grouping yields faster search results improving your application scalability.

Moreover, using groups to cache frequently accessed data falling under specific search criteria saves the cost of searching the database every time.

Properties

  • One - level hierarchy: There is no further level of hierarchy after the group. Cache items can be logically separated by a single group only.

  • Case Sensitive: Groups are case-sensitive.

  • String DataType: Groups can only be of string data type.

  • No Overlapping: Groups do not overlap and hold distinct names.

CRUD Operations with Groups

Ncache enables you to perform CRUD operations on data with group information by using CacheItem class. CacheItem lets you set additional specifications associated with an object as a property of Groups. The relevant details and behavior are explained below:

Add/Update Data with Groups

You can add or update a CacheItem in the cache by specifying a group. Meanwhile, an item can belong to a unique group while a group can have many associated items.

The Add operation adds a CacheItem with a group to the cache and fails with an exception if the key already exists in the cache. You can update the group of an existing item using an Insert operation that overwrites the previous group against the specified item. If the key or group against the key doesn't exist in the cache, the item is added with the group.

Besides, you can also add/insert a collection of items with groups to the cache using the AddBulk/InsertBulk methods.

Note

Using the Insert method is a recommended approach while adding an item with the group to the cache since it is fail-safe.

You can add or update a CacheItem in the cache with groups. In the case of add, if the group already exists, the operation will fail. In the case of update, if the group doesn't exist in the cache, it will be added to the cache, otherwise, it will overwrite the previous group.

Behavior

There can only be a single item associated with a single group. Meaning, you can't assign multiple groups to a CacheItem.

Retrieve Data with Groups

Once the items are added with groups to the cache, you can retrieve items from the cache by specifying the group as your search criteria. You can retrieve a list of keys and a dictionary of keys with cache items using GetGroupKeys and GetGroupData methods, respectively.

You can retrieve a bulk of CacheItem from the cache using groups. You can either retrieve a list of keys or a dictionary of keys and items using the GetGroupKeys() and GetGroupData() methods respectively.

Remove Data with Groups

You can remove a bulk of CacheItem from the cache using groups. You specify a group using the RemoveGroupData() method and then all of the keys and metadata that fall under the specified group are removed from the cache.

Client Cache Behavior

Write operations such as Add and update are first performed on the cluster cache and replicated to the client cache respectively. Read operations are directly performed on the server nodes.

In this Section

Add/Update Cache Data with Groups
Explains how an item can be added and updated to a data group.

Retrieve Cache Data with Groups
Explains how keys of a data group can be fetched from the cache.

Remove Cache Data with Group
Explains how a data group can be removed from the cache.