Three-site active federated deployment with ADO-enabled buckets
NOTE Do not delete buckets during rejoin after TSO. Delete operation will not be performed during rejoin execution.
When more than two sites are part of a replication group, and if network connectivity is interrupted between one site and the other two, and then write/update/ownership operations continue as they would with two sites, but the process for responding to read requests is more complex.
If an application requests an object that is owned by a site that is not reachable, ECS sends the request to the site with the secondary copy of the object. The secondary copy might have been subject to a data contraction operation, which is an XOR between two different datasets that produces a new dataset. The site with the secondary copy must retrieve the chunks of the object that is included in the original XOR operation, and it must XOR those chunks with the recovery copy. This operation returns the contents of the chunk that is originally stored on the owner site. Then the chunks from the recovered object can be reassembled and returned. When the chunks are reconstructed, they are also cached so that the site can respond more quickly to subsequent requests. Reconstruction is time consuming. More sites in a replication group imply more chunks that must be retrieved from other sites, and hence reconstructing the object takes longer. The following illustration shows the process for responding to read requests in a three-site federation.
Figure 1. Read request workflow example during a TSO in a three-site federation
Data is not available for the Topic
Please provide ratings (1-5 stars).
Please provide ratings (1-5 stars).
Please provide ratings (1-5 stars).
Please select whether the article was helpful or not.
Comments cannot contain these special characters: <>()\