A Cloud Storage Pool lets you use ILM to move object data outside of your StorageGRID system. For example, you might want to move infrequently accessed objects to low-cost Amazon Glacier storage, or you might want to free up on-premise storage by storing older versions of objects externally.
The following table compares storage pools to Cloud Storage Pools and shows the high-level similarities and differences.
Storage pool | Cloud Storage Pool | |
---|---|---|
How is it created? |
Using the Grid Manager. option inYou must set up storage grades before you can create the storage pool. |
Using the Grid Manager. option inYou must set up the external S3 bucket before you can create the Cloud Storage Pool. |
How many pools can you create? |
Multiple |
Only one |
Where are objects stored? |
On one or more Storage Nodes or Archive Nodes within StorageGRID. |
In a single S3 bucket that is external to the StorageGRID system. Note: Optionally, you can configure a bucket lifecycle for the external bucket if you want to transition objects to low-cost, long-term storage, such as Amazon Glacier. The external storage system must support the Glacier storage class and the S3 POST Object restore API.
|
What controls object placement? |
An ILM rule in the active ILM policy. |
An ILM rule in the active ILM policy. |
What data protection method is used? |
Replication |
Replication |
How many copies are allowed? |
Multiple |
One The replicated copy is moved to the Cloud Storage Pool and stored externally to StorageGRID. Additional replicated or erasure coded copies of the object cannot exist in StorageGRID during the same time period. |
What are the advantages? |
You can quickly access the object at any time. Redundant copies are available if a Storage Node or site fails. |
Low-cost storage. Note: Redundant copies are not available locally, and restore operations might take longer. For these reasons, a Cloud Storage Pool is an appropriate option for rarely accessed or older versions of data.
|