Abstract: A system can register a first client and a second client to respectively receive information about updates to a write-ahead log structured storage engine that comprises a log. The system can read an entry in the log, the entry being of an update type. The system can, in response to determining that the first client is registered to receive at least some of the information about updates that are of the update type, sending, to the first client, entry information about the entry. The system can, in response to determining that the second client is registered to receive information about updates that are of the update type, sending, to the second client, the entry information about the entry.

Abstract: The described technology is generally directed towards bulk-mode replication of data objects to a destination replication site. When replicating a group of objects, and there is enough object data to fill one or more chunk data structures, bulk-mode data replication is performed, with the destination site replicating the objects into chunk data structures and performing erasure coding to protect the data. This avoids the need for inefficient preliminary data protection scheme at the replication destination. When there are multiple objects to be replicated but not enough object data to fill a chunk data structure, bulk-mode data replication transfers the objects to the destination site for replicating as individual objects. Standard object-level replication can be used for single object replication. The technology thus facilitates high-performance data transfer to destination replication sites and efficient handling of data at the replication destination site.

Abstract: A system can register a first client and a second client to respectively receive information about updates to a write-ahead log structured storage engine that comprises a log. The system can read an entry in the log, the entry being of an update type. The system can, in response to determining that the first client is registered to receive at least some of the information about updates that are of the update type, sending, to the first client, entry information about the entry. The system can, in response to determining that the second client is registered to receive information about updates that are of the update type, sending, to the second client, the entry information about the entry.

Abstract: The described technology is generally directed towards quasi-listing table entries that facilitate the efficient return of an object list in response to a query for relevant objects in a data storage system. A quasi-listing table entry is created when an object is created, and subsequently converted to a normal listing table entry. A normal listing table entry is converted to a quasi-listing table entry when an object is deleted, with the quasi-listing table entry subsequently deleted. When an object list request is received, a normal listing table entry can be used directly to add to the object list; only if a quasi-listing table entry is encountered (before it can be converted or deleted) does the object table need to be accessed to determine if the object actually exists. This eliminates most object table accesses, including those in remote geographic zones, thereby making object listing significantly more efficient.

Abstract: The described technology is generally directed towards quasi-listing table entries that facilitate the efficient return of an object list in response to a query for relevant objects in a data storage system. A quasi-listing table entry is created when an object is created, and subsequently converted to a normal listing table entry. A normal listing table entry is converted to a quasi-listing table entry when an object is deleted, with the quasi-listing table entry subsequently deleted. When an object list request is received, a normal listing table entry can be used directly to add to the object list; only if a quasi-listing table entry is encountered (before it can be converted or deleted) does the object table need to be accessed to determine if the object actually exists. This eliminates most object table accesses, including those in remote geographic zones, thereby making object listing significantly more efficient.

Abstract: The described technology is generally directed towards bulk-mode replication of data objects to a destination replication site. When replicating a group of objects, and there is enough object data to fill one or more chunk data structures, bulk-mode data replication is performed, with the destination site replicating the objects into chunk data structures and performing erasure coding to protect the data. This avoids the need for inefficient preliminary data protection scheme at the replication destination. When there are multiple objects to be replicated but not enough object data to fill a chunk data structure, bulk-mode data replication transfers the objects to the destination site for replicating as individual objects. Standard object-level replication can be used for single object replication. The technology thus facilitates high-performance data transfer to destination replication sites and efficient handling of data at the replication destination site.

Abstract: The described technology is generally directed towards quasi-listing table entries that facilitate the efficient return of an object list in response to a query for relevant objects in a data storage system. A quasi-listing table entry is created when an object is created, and subsequently converted to a normal listing table entry. A normal listing table entry is converted to a quasi-listing table entry when an object is deleted, with the quasi-listing table entry subsequently deleted. When an object list request is received, a normal listing table entry can be used directly to add to the object list; only if a quasi-listing table entry is encountered (before it can be converted or deleted) does the object table need to be accessed to determine if the object actually exists. This eliminates most object table accesses, including those in remote geographic zones, thereby making object listing significantly more efficient.

Abstract: The described technology is generally directed towards quasi-listing table entries that facilitate the efficient return of an object list in response to a query for relevant objects in a data storage system. A quasi-listing table entry is created when an object is created, and subsequently converted to a normal listing table entry. A normal listing table entry is converted to a quasi-listing table entry when an object is deleted, with the quasi-listing table entry subsequently deleted. When an object list request is received, a normal listing table entry can be used directly to add to the object list; only if a quasi-listing table entry is encountered (before it can be converted or deleted) does the object table need to be accessed to determine if the object actually exists. This eliminates most object table accesses, including those in remote geographic zones, thereby making object listing significantly more efficient.

Abstract: Elastic cloud storage (ECS) systems typically divide storage nodes into geographic or topological zones and implement various concepts that enable the system to be extremely efficient in terms of capacity management. Architectures detailed herein can improve ECS and other similar systems in terms of inter-zone data transfers and remote data caching without violating core concepts of an ECS system that enable efficient capacity management.

Abstract: Elastic cloud storage (ECS) systems typically divide storage nodes into geographic or topological zones and implement various concepts that enable the system to be extremely efficient in terms of capacity management. Architectures detailed herein can improve ECS and other similar systems in terms of inter-zone data transfers and remote data caching without violating core concepts of an ECS system that enable efficient capacity management.