Abstract: A method for deleting a set of keys from a storage server is provided. The method includes generating a probabilistic data structure for a first set of keys and for each key in a second set of keys, determining whether a key of the second set of keys is found in the probabilistic data structure. The method includes identifying the key as a candidate for deletion if the key is not found in the probabilistic data structure. A system is also provided.

Abstract: Utilizing different data compression algorithms based on characteristics of a storage system, including: selecting, in dependence upon a priority for conserving processing resources or storage resources in a storage system, a data compression algorithm to utilize to compress data; detecting that at least one of an amount of processing resources available in the storage system or the amount of space available to store additional data in the storage system has changed; and responsive to detecting that at least one of the amount of processing resources available in the storage system or the amount of space available to store additional data in the storage system has changed, selecting a different data compression algorithm to utilize to compress data.

Abstract: A system and method for managing tables in a storage system is described.

Abstract: A system and method comprising: receiving a request to write data stored at a first range of a first volume to a second range of a second volume, where first metadata for the first range of the first volume is associated with a range of physical addresses where the data is stored in the storage system; and responsive to receiving the request: creating second metadata for the second range of the second volume, wherein the second metadata is associated with the range of physical addresses where the data is stored in the storage system; and associating the second volume with the second metadata.

Abstract: A method of cache management, the method comprising: identifying, among a plurality of storage items, storage items having an access count above a first threshold to generate a set of storage items; identifying, among the set of storage items, storage items having an updated access count above a second threshold to generate a subset of storage items, wherein, for each storage item, the updated access count is dependent upon a number of accesses subsequent to generating the set of storage items; and adding the storage items of the subset of storage items to a cache.

Abstract: Restoring lost data including detecting that a portion of the dataset stored in a first storage system has become unavailable, obtaining an identifier for the portion of the dataset, locating, using the identifier, a replacement portion of the dataset that is stored at one or more other storage systems, and writing, to the dataset that is stored in the first storage system, the replacement portion of the dataset as a replacement of the portion of the dataset that has become unavailable, where the writing occurs automatically, without user intervention.

Abstract: Methods, computer systems, and computer readable medium are described for low latency reads using cached deduplicated data, including: receiving a request to read data from a storage system; query, using a generated hash value associated with the request to read data, one or more deduplication tables that corresponds to the hash value; and responsive to determining that the one or more deduplication tables includes an entry that corresponds to the hash value, using a mapping contained in the entry to perform the requested to read data, wherein the mapping includes a pointer to a physical location where at least a portion of the data is stored.

Abstract: Sharing encryption information amongst storage devices in a storage system, including: detecting that one or more source storage devices has been designated for removal from the storage system; and responsive to detecting that the source storage devices have been designated for removal from the storage system: encrypting a device key for each remaining storage device using a master secret to generate an encrypted device key for each remaining storage device; generating, for each write group, a first plurality of shares from the master secret; creating, from each share of the first plurality of shares, a second plurality of shares equal to a minimum number of the remaining storage devices required from each write group for a quorum to boot the storage system; and storing, for each remaining storage device, the encrypted device key and a respective share of the second plurality of shares in the remaining storage devices.

Abstract: Intra-block error correction including generating reconstructed sub-portions of a portion of a dataset that is distributed into a plurality of portions that are stored on different storage devices of the storage system, wherein the dataset indicates a first number of errors that is greater than a second number of parity data portions of the dataset, determining that a sub-portion of the dataset does not match a corresponding reconstructed sub-portion of the dataset; and based on the determination, replacing the sub-portion of the dataset with the reconstructed sub-portion of the dataset.

Abstract: The disclosure relates generally to devices and methods for treating a tissue defect, for example, by suturing. In some embodiments, a suturing device may include an elongate member having a working channel, a suture channel, and a suture arm extending from the elongate member. The suturing device may further include a needle passer located within the working channel, the needle passer operable to deliver a needle between the elongate member and a distal end of the suture arm for suturing a target tissue, and a suture extending through the suture channel, wherein the suture is coupled to the needle. The suturing device may further include a plurality of imaging devices, wherein a first imaging device is positioned along a distal face of the elongate member, and wherein a second imaging device is positioned along the suture arm.

Abstract: Techniques for selectively offloading data that is computed by a first processing unit during training of an artificial neural network onto memory associated with a second processing unit and transferring the data back to the first processing unit when the data is needed for further processing are described herein. For example, the first processing unit may compute activations for operations associated with forward propagation. During the forward propagation, one or more of the activations may be transferred to a second processing unit for storage. Then, during backpropagation for the artificial neural network, the activations may be transferred back to the first processing unit as needed to compute gradients.

Abstract: In a storage system that includes a plurality of storage devices configured into one or more write groups, quorum-aware secret sharing may include: encrypting a device key for each storage device using a master secret; generating a plurality of shares from the master secret such that a minimum number of storage devices required from each write group for a quorum to boot the storage system is not less than a minimum number of shares required to reconstruct the master secret; and storing the encrypted device key and a separate share of the plurality of shares in each storage device.

Abstract: Self-tuning clusters for resilient microservices, including: receiving, by a services orchestrator within a cloud-computing environment and from a plurality of cloud computing instances, respective latency measurements corresponding to respective control plane operations directed to a shared resource of the plurality of cloud computing instances; determining, based on a current timeout value and on the respective latency measurements from the plurality of cloud computing instances, an updated timeout value for the shared resource; and providing, to each of the plurality of cloud computing instances, the updated timeout value for the shared resource.

Abstract: A system and method for offset protection data in a RAID array. A computer system comprises client computers and data storage arrays coupled to one another via a network. A data storage array utilizes solid-state drives and Flash memory cells for data storage. A storage controller within a data storage array is configured to store user data in a first page of a first storage device of the plurality of storage devices; generate intra-device protection data corresponding to the user data, and store the intra-device protection data at a first offset within the first page. The controller is further configured to generate inter-device protection data corresponding to the first page, and store the inter-device protection data at a second offset within a second page in a second storage device of the plurality of storage devices, wherein the first offset is different from the second offset.

Abstract: A method for deleting a set of keys from a storage server is provided. The method includes generating a probabilistic data structure for a first set of keys and for each key in a second set of keys, determining whether a key of the second set of keys is found in the probabilistic data structure. The method includes identifying the key as a candidate for deletion if the key is not found in the probabilistic data structure. A system is also provided.

Abstract: A system and method comprising: receiving a request to write data stored at a first range of a first volume to a second range of a second volume, where first metadata for the first range of the first volume is associated with a range of physical addresses where the data is stored in the storage system; and responsive to receiving the request: creating second metadata for the second range of the second volume, wherein the second metadata is associated with the range of physical addresses where the data is stored in the storage system; and associating the second volume with the second metadata.

Abstract: Techniques for determining unified futures of objects in an environment are discussed herein. Techniques may include determining a first feature associated with an object in an environment and a second feature associated with the environment and based on a position of the object in the environment, updating a graph neural network (GNN) to encode the first feature and second feature into a graph node representing the object and encode relative positions of additional objects in the environment into one or more edges attached to the node. The GNN may be decoded to determine a distribution of predicted positions for the object in the future that meet a criterion, allowing for more efficient sampling. A predicted position of the object in the future may be determined by sampling from the distribution.

Abstract: Techniques for determining unified futures of objects in an environment are discussed herein. Techniques may include determining a first feature associated with an object in an environment and a second feature associated with the environment and based on a position of the object in the environment, updating a graph neural network (GNN) to encode the first feature and second feature into a graph node representing the object and encode relative positions of additional objects in the environment into one or more edges attached to the node. The GNN may be decoded to determine a predicted position of the object at a subsequent timestep. Further, a predicted trajectory of the object may be determined using predicted positions of the object at various timesteps.

Abstract: Techniques for determining unified futures of objects in an environment are discussed herein. Techniques may include determining a first feature associated with an object in an environment and a second feature associated with the environment and based on a position of the object in the environment, updating a graph neural network (GNN) to encode the first feature and second feature into a graph node representing the object and encode relative positions of additional objects in the environment into one or more edges attached to the node. The GNN may be decoded to determine a first predicted position of the object. The first predicted position may be determined to be outside of a bounded area of the environment. Based on this determination, a second predicted position of the object may be determined using map data associated with the object.

Abstract: A system and method for managing tables in a storage system is described.