Abstract: The present disclosure relates to capturing snapshots of key management data and storing the snapshots for efficient re-creation of the key management data in the event of an outage at one or more nodes. A snapshot orchestrator can request snapshot instances from each of the series of nodes across one or more regions in a cloud infrastructure service. Each snapshot instance can provide a plurality of modifications to a plurality of client keys maintained by each of the series of nodes. The snapshot orchestrator can obtain the snapshot instances and corresponding metadata and verify the snapshot instances received from the series of nodes. The snapshot orchestrator can store the verified snapshot instance and corresponding metadata at a storage node. This can allow for subsequent retrieval of the snapshot instance and recreating the log record and modifications to the plurality of keys at any of the series of nodes.

Abstract: Discussed herein is a technique for replication of keys across regions of a cloud infrastructure. A first vault is created in a first region. The first vault stores a plurality of records, each of which is associated with a key and corresponding metadata. A second region, different from the first region, is selected where replication of the plurality of records is desired. A second vault is created in the second region. The plurality of records are relayed from the first vault to the second vault, so that each of the plurality of records is replicated in the second vault based on an entropy value of the record. A mutation operation executed with respect to the first vault is stored as a new record in the first vault. The new record is transmitted to the second vault, which is updated to reflect the mutation operation performed on the first vault.

Abstract: Discussed herein is a technique for replication of keys across regions of a cloud infrastructure. A first vault is created in a first region. The first vault stores a plurality of records, each of which is associated with a key and corresponding metadata. A second region, different from the first region, is selected where replication of the plurality of records is desired. A second vault is created in the second region. The plurality of records are relayed from the first vault to the second vault, so that each of the plurality of records is replicated in the second vault based on an entropy value of the record. A mutation operation executed with respect to the first vault is stored as a new record in the first vault. The new record is transmitted to the second vault, which is updated to reflect the mutation operation performed on the first vault.

Abstract: Discussed herein is a technique for replication of keys across regions of a cloud infrastructure. A first vault is created in a first region. The first vault stores a plurality of records, each of which is associated with a key and corresponding metadata. A second region, different from the first region, is selected where replication of the plurality of records is desired. A second vault is created in the second region. The plurality of records are relayed from the first vault to the second vault, so that each of the plurality of records is replicated in the second vault based on an entropy value of the record. A mutation operation executed with respect to the first vault is stored as a new record in the first vault. The new record is transmitted to the second vault, which is updated to reflect the mutation operation performed on the first vault.

Abstract: The present disclosure relates to capturing snapshots of key management data and storing the snapshots for efficient re-creation of the key management data in the event of an outage at one or more nodes. A snapshot orchestrator can request snapshot instances from each of the series of nodes across one or more regions in a cloud infrastructure service. Each snapshot instance can provide a plurality of modifications to a plurality of client keys maintained by each of the series of nodes. The snapshot orchestrator can obtain the snapshot instances and corresponding metadata and verify the snapshot instances received from the series of nodes. The snapshot orchestrator can store the verified snapshot instance and corresponding metadata at a storage node. This can allow for subsequent retrieval of the snapshot instance and recreating the log record and modifications to the plurality of keys at any of the series of nodes.

Abstract: Discussed herein is a technique for replication of keys across regions of a cloud infrastructure. A first vault is created in a first region. The first vault stores a plurality of records, each of which is associated with a key and corresponding metadata. A second region, different from the first region, is selected where replication of the plurality of records is desired. A second vault is created in the second region. The plurality of records are relayed from the first vault to the second vault, so that each of the plurality of records is replicated in the second vault based on an entropy value of the record. A mutation operation executed with respect to the first vault is stored as a new record in the first vault. The new record is transmitted to the second vault, which is updated to reflect the mutation operation performed on the first vault.