Abstract: Techniques for training a codebook usable by a machine learned model to predict an object trajectory or scene data are described herein. For example, the techniques may include generating tokens representing discrete object behavior into a machine learned model that outputs a sequence of tokens that is usable by another machine learned model to generate the object trajectory (e.g., position data, velocity data, acceleration data, etc.) or the scene data associated with the environment. The object trajectory can be sent to a vehicle computing device for consideration during vehicle planning, which may include simulation.

Abstract: Techniques for predicting an object trajectory or scene information are described herein. For example, the techniques may include inputting tokens representing discrete behavior into a machine learned model. The machine learned model may output a sequence of tokens that is usable by another machine learned model to generate an object trajectory (e.g., position data, velocity data, acceleration data, etc.) for one or more objects in the environment. The object trajectory can be sent to a vehicle computing device for consideration during vehicle planning, which may include simulation.

Abstract: Techniques for predicting an object trajectory or scene information are described herein. For example, the techniques may include inputting latent variable data into a machine learned model. The machine learned model may output an object trajectory (e.g., position data, velocity data, acceleration data, etc.) for one or more objects in the environment based on the latent variable data. The object trajectory can be sent to a vehicle computing device for consideration during vehicle planning, which may include simulation.

Abstract: Providing authorization and authentication in a cloud for a user of a storage array includes: receiving, by a storage array access module from a client-side array services module, a token representing authentication of user credentials and authorized access privileges defining one or more storage array services accessible by the user, where the token is generated by a cloud-based security module upon authentication of the user credentials and identification of authorized access privileges for the user; receiving, by the storage array access module from the user, a user access request to one or more storage array services; and determining, by the storage array access module, whether to grant the user access request in dependence upon the authorized access privileges represented by the token.

Abstract: Data protection in a storage system that includes a plurality of Non-Volatile Memory Express (‘NVMe’) Solid State Drives (‘SSDs’), including: retrieving, from a plurality of NVMe SSDs (‘Non-Volatile Memory Express Solid State Drives’) of a storage system, one or more unencrypted shares of a master secret; reconstructing the master secret using the shares of the master secret; decrypting one or more encrypted device keys using the master secret; and using the decrypted device keys to perform a plurality of accesses to one or more of the NVMe SSDs.

Abstract: An apparatus, method, and computer-readable storage medium for allowing a block-addressable storage device to provide a sparse address space to a host computer. The storage device exports an address space to a host computing device which is larger than the storage capacity of the storage device. The storage device translates received file system object addresses in the larger address space to physical locations in the smaller address space of the storage device. This allows the host computing device more flexibility in selecting addresses for file system objects which are stored on the storage device.

Abstract: Techniques for increasing performance of machine-learned models while conserving computational resources generally required by ensemble machine-learning methods are described herein. The techniques may include determining multiple views of a scene that is to be input into a machine-learned model. In some examples, a scene data input may be rotated by 90, 180, and 270 degrees to generate four scene inputs (e.g., 0-, 90-, 180-, and 270-degree rotated inputs) that can be passed through the machine-learned model and the results per scene can be aggregated to determine a final prediction/decision. Similarly, scene inputs may be shifted, reflected, translated, and/or the like before being input into the machine-learned model. The predictions may be associated with one or more objects in the environment that are represented in the scenes.

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: Prioritizing garbage collection based on the extent to which data is deduplicated, including: determining, for one or more data elements, a number of deduplicated references to each data element; storing, for each of the data elements, the data element in an area of the storage device that contains other data elements with a similar number of deduplicated references; and adjusting a garbage collection schedule for the storage device, wherein garbage collection operations are performed more frequently on areas of the storage device that contain data elements with a relatively low number of deduplicated references.

Abstract: Techniques for improving operational decisions of an autonomous vehicle are discussed herein. In some cases, a system may generate reference graphs associated with a route of the autonomous vehicle. Such reference graphs can comprise precomputed feature vectors based on grid regions and/or lane segments. The feature vectors are usable to determine scene context data associated with static objects to reduce computational expenses and compute time.

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: A method including, responsive to receiving a request identifying a volume and indicating a command to take a snapshot of the volume, mapping a second logical grouping of data to reference the first logical grouping of data, and remapping the first volume to map to the second logical grouping of data instead of the first logical grouping of data such that the first volume remains addressable with similar access permissions before and after creating the snapshot. The method also includes, in response to receiving a write request targeting the second logical grouping, splitting the second logical grouping into a plurality of ranges including a first range and a second range; wherein the first range of the second logical grouping maps to the first logical grouping, and the write request is performed on the second range of the second logical grouping.

Abstract: Utilizing different data compression algorithms based on characteristics of a storage system, including: detecting that an availability of resources to a storage system for storing additional data in the storage system has changed; and based on the detection, selecting a data compression algorithm to utilize to compress data that is different compared to a previously selected data compression algorithm.

Abstract: A system and method for performing replication using mediums. A snapshot ‘S’ is selected for replication, and the anchor medium ‘M’ of S is identified. The replica storage array ‘R’ to which M is being replicated determines a list of medium extents which are available. R sends this list to an original storage array ‘O’ and O generates a list of medium extents which need to be sent to R. Only those medium extents corresponding to M which R does not already have will be sent from O to R. Also, R can obtain medium extents from other source storage arrays during the replication process.

Abstract: Providing for increased flexibility for large scale parity, the including: writing data to a storage system, including utilizing a first data protection scheme; identifying, for storage media in the storage system, characteristics of the storage media; identifying, in dependence up the characteristics for the storage media, a second data protection scheme to use for the data; and writing the data to the to the storage system utilizing the second data protection scheme.

Abstract: A method including, based on a request targeting a logical volume mapped to a logical grouping of data, splitting the logical grouping of data into two or more ranges including a first range and a second range, wherein lookups corresponding to one or more blocks targeted by the request are performed in the first range of the two or more ranges, and performing the request using the second range.

Abstract: A system, method, and product for flexible RAID layouts in a storage system, including: determining a reliability of an individual storage device of a plurality of storage devices, the individual storage device containing a plurality of portions of a Redundant Array of Independent Disks (RAID) stripe in a storage system, wherein the RAID stripe includes user data and inter-device parity data; detecting a change in the reliability of the individual storage device that contains the portion corresponding to the RAID stripe; and changing an amount of intra-device protection corresponding to the RAID stripe by decreasing, in the RAID stripe, an amount of space used to store the inter-device protection data.

Abstract: Intelligent responses to errors in a storage system, including: after a first attempt to read data from a first set of resources in a storage system results in an error, determining whether to issue a second attempt to read data from the first set of resources in a storage system; responsive to determining not to issue the second attempt to read data from the first set of resources in a storage system, retrieving the data from a second set of resources in the storage system; and responsive to determining to issue the second attempt to read data from the first set of resources in a storage system, issuing a second read attempt to read the data, wherein the error correction effort level associated with the second attempt is increased relative to the error correction effort level associated with the first attempt.