A plurality of computing devices are communicatively coupled to each other via a network, and each of the plurality of computing devices is operably coupled to one or more of a plurality of storage devices. A plurality of failure resilient stripes is distributed across the plurality of storage devices such that each of the plurality of failure resilient stripes spans a plurality of the storage devices. A graphics processing unit is operable to access data files from the failure resilient stripes, while bypassing a kernel page cache. Furthermore, these data files may be accessed in parallel by the graphics processing unit.

A method for execution by a distributed storage (DS) unit of a dispersed storage network (DSN), includes receiving a set of write slice requests that includes a set of slice names that are not utilized to store encoded data slices (EDSs), where each write slice request from the set of slice names that are not utilized to store encoded data slices includes a trap slice. The method continues with an access slice request that includes a requested slice name being received from a second DSN client, and then continues by determining whether the requested slice name corresponds to a trap slice stored by the DS unit. When the requested slice name corresponds to a trap slice, an action is determined for the access slice request based on an anomaly processing scheme.

A method begins with a processing module selecting one of a plurality of dispersed storage (DS) processing modules for facilitating access to a dispersed storage network (DSN) memory. The method continues with the processing module sending a DSN memory access request to the one of the plurality of DS processing modules. The method continues with the processing module selecting another one of the plurality of DS processing modules when no response is received within a given time frame or when the response to the access request does not include an access indication. The method continues with the processing module sending the DSN memory access request to the another one of the plurality of DS processing modules.

To identify slice errors, a processing module of a computing device in a dispersed storage network (DSN) sends first list digest requests to at least first and second dispersed storage (DS) units. The requests indicates a first range of slice names to include in a first list digest. The processing module receives digest responses from the DS units, and compares the digest responses to determine whether they identify the same slices. If they do not identify the same slices, the processing module sends second list digest requests indicating a sub-range of the first range of slice names to include in second list digests. The sub-range continues to be narrowed until the processing module identifies at least one sub-range of slice names where a slice error exists.

A method, system, and computer program product for implementing indexes in a dispersed storage network (dsNet) are provided. The method receives a key-value list request including a start key. In response to the key-value list request, a set of key-value list requests are transmitted to a plurality of data source units within a dispersed storage network (dsNet). The method identifies a set of keys returned from the plurality of data source units and identifies a subset of keys from the set of keys returned. The subset of keys including a union of the set of keys. The key-value pairs associated with the subset of keys are restored. The method generates a key-value list response to the key-value list request. The key-value list response includes the restored key-value pairs associated with the subset of keys.

A method includes a computing device of a storage network dispersed storage error encoding a plurality of data segments to produce a plurality of sets of encoded data slices. The method further includes the computing device obfuscating a first set of encoded data slices of the plurality of sets of encoded data slices using an obfuscating method to produce a first set of obfuscated encoded data slices. The method further includes the computing device outputting the first set of obfuscated encoded data slices for storage in the storage network.

A data object to be stored in one or more memories as a first set of encoded data segments is received at a storage processing module. An index to include an entry identifying the data object to be stored is updated, and storage of the data object is initiated. Initiating storage of the data object includes initiating storage of the first set of encoded data segments. A determination is made regarding whether storage of the data object to the one or more memories has been completed. If storage of the data object has been completed, the entry identifying the data object to be stored is removed from the index.

Methods and apparatus for selection of memory devices in a distributed storage network. In an embodiment, a computing device receives a data object for storage and forwards the data object to a buffer for temporary storage, the buffer comprised of a first memory devices of a first memory type. A system level storage efficiency is determined for the data object based, at least in part, on a data attribute associated with the data object. Second memory devices, of a second memory type, are selected based on the system level storage efficiency preference, and compatible dispersed storage error encoding parameters for the data object are determined. The data object is encoded using the encoding parameters to generate a plurality of encoded data slices, which are provided to the second plurality of memory devices for storage. Further, system addressing information is generated based on an identifier associated with the data object.

A method for rebuilding data in a distributed storage network (DSN) including a plurality of storage units. A rebuilding module of the DSN selects an address range associated with memory of a storage unit of the DSN, and initiates execution of data rebuilding activities for the selected address range. The rebuilding module determines a first rebuilding rate of the storage unit for a timeframe subsequent to initiating the execution of the data rebuilding activities. The rebuilding module further operates to pause the execution of the data rebuilding activities for a pause timeframe, and determine a second rebuilding rate of the storage unit for the pause timeframe. In response to determining that the second rebuilding rate compares favorably to the first rebuilding rate, the rebuilding module resumes execution of the data rebuilding activities. If the comparison is unfavorable, a second address range is selected for data rebuilding activities.

A plurality of computing devices are communicatively coupled to each other via a network, and each of the plurality of computing devices is operably coupled to one or more of a plurality of storage devices. A plurality of failure resilient address spaces are distributed across the plurality of storage devices such that each of the plurality of failure resilient address spaces spans a plurality of the storage devices. The plurality of computing devices maintains metadata that maps each failure resilient address space to one of the plurality of computing devices. The metadata is grouped into buckets. Each bucket is stored in the backend of a computing device. Data may be migrated from an external file system to the plurality of storage devices using inode stubs to represent directories and files of the external file system. As the contents of the external file system are copied, the inode stubs are replaced with real inodes.