Abstract: According to one set of embodiments, a computer system can receive a request or command to delete a snapshot from among a plurality of snapshots of a dataset, where the plurality of snapshots are stored in cloud/object storage. In response, the computer system can add the snapshot to a batch of pending snapshots to be deleted and can determine whether the size of the batch has reached a threshold. If the size of the batch has not reached the threshold, the computer system return a response to an originator of the request or command indicating that the snapshot has been deleted, without actually deleting the snapshot from the cloud/object storage.

Abstract: Techniques for archiving data in cloud/object storage using local metadata staging are provided. In one set of embodiments, a computer system residing at an on-premises site comprising on-premises storage can receive a snapshot of a dataset to be archived. The computer system can package data in the snapshot into one or more fixed-size data chunks and upload the one or more fixed-size data chunks to cloud/object storage. Further, concurrently with the packaging and the uploading, the computer system can stage metadata for the snapshot in the on-premises storage. Then, upon uploading all of the data of the snapshot, the computer system can upload the metadata staged in the on-premises storage to the cloud/object storage.

Abstract: Techniques for managing overwrites when archiving data in cloud/object storage are provided. In one set of embodiments, a computer system can overwrite a chunk of data residing in cloud/object storage, where the cloud/object storage only guarantees eventual consistency for overwrites, and where the overwriting is associated with a sequence number. Upon overwriting the chunk of data, the computer system create a new instance of a file in the cloud/object storage, where the new instance of the file has a version number corresponding to the sequence number.

Abstract: According to one set of embodiments, a computer system can receive a request/command to delete a snapshot from among a plurality of snapshots of a dataset that are stored in cloud/object storage. In response, the computer system can determine a first minimum chunk identifier (ID) and a first maximum chunk ID of a range of data chunks referred to by the first snapshot, and a second minimum chunk ID and a second maximum chunk ID of a range of data chunks referred to by an immediate child snapshot of the first snapshot. The computer system can then, for each data chunk identified by a chunk ID spanning from the second minimum chunk ID to the first maximum chunk ID, check whether the data chunk is referred to by the immediate child snapshot and if not, delete the data chunk from the cloud/object storage.

Abstract: According to one set of embodiments, a computer system can receive a request/command to delete a snapshot from among a plurality of snapshots of a dataset that are stored in cloud/object storage. In response, the computer system can determine a first minimum chunk identifier (ID) and a first maximum chunk ID of a range of data chunks referred to by the first snapshot, and a second minimum chunk ID and a second maximum chunk ID of a range of data chunks referred to by an immediate child snapshot of the first snapshot. The computer system can then, for each data chunk identified by a chunk ID spanning from the second minimum chunk ID to the first maximum chunk ID, check whether the data chunk is referred to by the immediate child snapshot and if not, delete the data chunk from the cloud/object storage.

Abstract: Techniques for deleting a snapshot of a data set archived in cloud/object storage are provided. In one set of embodiments, a computer system can identify data blocks that are part of the snapshot and are referred to by an incremental child snapshot of the snapshot. The computer system can further create a new metadata object that includes metadata entries for the identified data blocks, each metadata entry comprising an identifier of the incremental child snapshot, an identifier of a data block, and an identifier of a data object of the incremental child snapshot where the data block is stored. The computer system can then associate the new metadata object with the incremental child snapshot in the cloud/object storage.

Abstract: Techniques for archiving a data set in cloud/object storage are provided. In one set of embodiments, a computer system can receive an incremental snapshot of the data set and package the incremental snapshot into one or more data objects and one or more metadata objects. The one or more data objects can include data blocks of the incremental snapshot, and the one or more metadata objects can include a metadata entry for each data block, the metadata entry comprising an identifier of the incremental snapshot, an identifier of the data block, and an identifier of a data object in which the data block is stored. The computer system can then write the one or more data objects and the one or more metadata objects to the cloud/object storage.

Abstract: A data storage device includes a cache for a data storage and a processor. The data storage includes an object storage. The processor obtains cache hardware heuristics data for a first time period; makes a first determination that the cache hardware heuristics data for the first time period does not meet a goal associated with the first time period; and populates the cache using a reduced size index cache in response to the first determination during a second time period.

Abstract: Smart weighted container data cache eviction preserves write evict units (WEUs) containing the most frequently and recently accessed blocks to maintain low latency data cache. Prior to performing cache eviction, the WEUs are weighted based on the page statistics maintained for each WEU. Page statistics include page hit/frequency and recency statistics associated with each WEU and data cache eviction is performed at the WEU level of granularity. Therefore, an entire WEU can be evicted based on page hit/frequency and recency statistics associated with the WEU.

Abstract: Techniques for handling fragmentation of archived data in cloud/object storage are provided. In one set of embodiments, a computer system can upload a new snapshot of a data set to the cloud/object storage, where the new snapshot comprises a plurality of data blocks, and where the new snapshot is uploaded as one or more data objects and one or more metadata objects. For each data block in the plurality of data blocks, the computer system can identify an existing data object in the cloud/object storage where the data block is currently stored. The computer system can further select, from among the identified existing data objects, a subset of the existing data objects that are part of a snapshot created in the cloud/object storage at least a predefined number of snapshots before the new snapshot, and select one or more data blocks of one or more data objects in the subset that have not be overwritten by another snapshot.

Abstract: Techniques for supporting glacier tiering of archived data in cloud/object storage are provided. In one set of embodiments, a computer system can identify one or more snapshots of a data set archived in the cloud/object storage that are scheduled to be migrated from a standard tier of the cloud/object storage to a glacier tier of the cloud/object storage. The computer system can further determine one or more data objects of the one or more snapshots that are referred to by at least one child snapshot and clone the one or more data objects in the standard tier.

Abstract: Techniques for deleting a snapshot of a data set archived in cloud/object storage are provided. In one set of embodiments, a computer system can identify data blocks that are part of the snapshot and are referred to by an incremental child snapshot of the snapshot. The computer system can further create a new metadata object that includes metadata entries for the identified data blocks, each metadata entry comprising an identifier of the incremental child snapshot, an identifier of a data block, and an identifier of a data object of the incremental child snapshot where the data block is stored. The computer system can then associate the new metadata object with the incremental child snapshot in the cloud/object storage.

Abstract: Techniques for archiving a data set in cloud/object storage are provided. In one set of embodiments, a computer system can receive an incremental snapshot of the data set and package the incremental snapshot into one or more data objects and one or more metadata objects. The one or more data objects can include data blocks of the incremental snapshot, and the one or more metadata objects can include a metadata entry for each data block, the metadata entry comprising an identifier of the incremental snapshot, an identifier of the data block, and an identifier of a data object in which the data block is stored. The computer system can then write the one or more data objects and the one or more metadata objects to the cloud/object storage.

Abstract: A data storage device includes a cache for an object storage and a processor. The processor suspends processing of files for storage in the object storage. While the processing of files is suspended the processor generates a rebuilt index using the object storage, generates a rebuilt index cache using the object storage, stores the rebuilt index in the object storage, and stores the rebuilt index cache in the cache.

Abstract: An adaptive smart data cache eviction method takes file-based quotas into account during eviction of WEUs as opposed to a default eviction policy that treats all files the same. Adaptive smart data cache eviction is a granular and dynamic eviction of the least frequently and least recently accessed blocks contained in a WEU for those files that have exceeded file-based quotas established for them or that are determined to be candidates for eviction based on the number of blocks stored for them relative to other files and how frequently and recently those blocks were accessed.

Abstract: An intelligent cache pre-fetch system includes a pre-fetch throttling scheme to monitor a cache hit rate context. Pre-fetch reads of additional data are only launched when the context is below a given threshold. A pre-fetch read of additional data can be selectively initiated after determining that references to neighboring segments related to a compression region already in memory are not yet present in the cache. Additional throttling of pre-fetch reads can be accomplished by only initiating the selective pre-fetch of additional data after determining whether the compression region to which the neighboring segments are related is a hot region, where a hot region is characterized as a compression region having data that is accessed frequently as compared to data in other compression regions.