What Is the Cloud Cache? – Implement and Manage FSLogix

What Is the Cloud Cache?

The cloud cache is an alternative to the profile container, and it also provides some additional functionality to the profile container. The cloud cache uses a local profile to service all reads from a redirected profile or office container, after the first read. The cloud cache also allows the use of multiple remote locations, which are all continually updated

during the user session. Using the cloud cache can insulate users from ­short-­term

loss of connectivity to remote profile containers. The cloud cache can also provide a real-time, “active-active” redundancy for the profile container and office container.

It’s important to understand that, even with cloud cache, all initial reads are accomplished from the redirected location. Likewise, all writes occur to all remote storage locations, although writes go to the local cache file first.

The cloud cache doesn’t improve the users’ sign-on and sign-out experience when using poorly performing storage. It’s common for environments using cloud cache to have slightly slower sign-on and sign-out times, relative to using traditional VHDLocations, using the same storage. After initial sign-on, the cloud cache can improve the user experience for subsequent reads of data from the profile container or office container, as these reads are serviced from the local cache file.

Cloud Cache Design and Functionality

The cloud cache uses one or more remote profile containers, along with metadata. The combination of a profile container and metadata is referred to as a cloud cache provider. The cloud cache uses a local cache file that contains part of the dataset stored in the cloud cache providers. The data gets stored and accessed from the local cache file after any read from a cloud cache provider. Additionally, the local cache file will service any writes from the system and then send those writes to all cloud cache providers in the cloud cache configuration registry. This is a synchronous process that depends on the performance of the various components such as client, network, and storage.

If a provider is not available, then the system will continue to operate with the remaining provider. If a provider that was unavailable becomes available before the user signs out, then it will be brought up-to-date from the local cache. When a provider isn’t available when the user signs out, it will be brought up-to-date in subsequent sessions by having all its data replaced from an existing and up-to-date provider. If all remote providers are stale, the provider with the latest metadata is considered the source of truth.