Technologies are provided to ensure integrity of erasure coded data that is subject to read and write access from distributed processes. Multiple processes that access erasure coded data can be coordinated in an efficient, scalable and fault-tolerant manner so that integrity of the original data is maintained. The Technologies include a fault-tolerant access coordination protocol that ensures exclusive write access by a client. The coordination protocol achieves scalability by not relying on centralized components, and achieves efficiency and performance by piggy-packing access coordination messages on operations of the underlying erasure coding protocol.

In a method disclosed for writing data, a device receives data, divides the data into one or more data fragments, obtains a first parity fragment based on the one or more data fragments and a second parity fragment of a written data fragment in a stripe distributed across a plurality of nodes, stores the one or more data fragments and the first parity fragment in the stripe.

Various embodiments set forth techniques for free space management in a block store. The techniques include receiving a request to allocate one or more blocks in a block store, accessing a sparse hierarchical data structure to identify an allocator page identifying a region of a backing store having a greatest number of free blocks, and allocating the one or more blocks.

A method for synchronous replication of stream data includes receiving a stream of data blocks for storage at a first storage location associated with a first geographical region and at a second storage location associated with a second geographical region. The method also includes synchronously writing the stream of data blocks to the first storage location and to the second storage location. While synchronously writing the stream of data blocks, the method includes determining an unrecoverable failure at the second storage location. The method also includes determining a failure point in the writing of the stream of data blocks that demarcates data blocks that were successfully written and not successfully written to the second storage location. The method also includes synchronously writing, starting at the failure point, the stream of data blocks to the first storage location and to a third storage location associated with a third geographical region.

A method for execution by a dispersed storage network (DSN) managing unit includes receiving access information from a plurality of distributed storage and task (DST) processing units via a network. Cache memory utilization data is generated based on the access information. Configuration instructions are generated for transmission via the network to the plurality of DST processing units based on the cache memory utilization data.

Software that may be implemented using a circuit is disclosed. The software may include an Application Programming Interface (API) to receive a request from an application relating to a key-value pair for a Key-Value Solid State Drive (KV-SSD). The key-value pair may include a key and a value; the application may be executed by a processor. The software may also include combiner software to combine the key with an index to produce an indexed key, and execution software to execute an operation on the KV-SSD using the indexed key and the value.

Techniques are disclosed for content storage in a way that facilitates consistent and concurrent read/write processing of stored documents. An example methodology implementing the techniques includes segmenting the contents of a document into a plurality of content segments and storing the plurality of content segments within a data structure, the data structure including storage blocks having storage portions and buffer portions. The storage of the plurality of content segments includes storage of content segments within a storage portions of the storage blocks of the data structure. The method also includes receiving at least one change to the content and utilizing a buffer portion of at least one storage block to store the at least one change to the content.

A control method for a solid state drive is provided. The solid state drive includes a non-volatile memory with plural blocks. In a step (a1), a block is opened. In a step (a2), a program action is performed to store a valid write data into the open block. Then, a step (a3) is performed to judge whether an amount of the valid write data in the open block reaches a predetermined capacity. In a step (a4), if the amount of the valid write data in the open block does not reach the predetermined capacity, the step (a2) is performed again. In a step (a5), if the amount of the valid write data in the open block reaches the predetermined capacity, the open block is closed and the step (a1) is performed again. The predetermined capacity is lower than a capacity of one block.

The present disclosure relates to method for checking the reading phase of a non-volatile memory device including at least an array of memory cells and with associated decoding and sensing circuitry and a memory controller, the method comprises: storing in a dummy row associated to said memory block at least internal block variables and a known pattern; performing a reading of said dummy row; comparing a result of the reading with the known pattern; trimming the parameters of the reading and/or swapping the used memory block based on the result of the comparing.

A data access method performed by a processing device is disclosed. The processing device manages a storage device based on a log-structured file system (LFS), where the processing device writes a target file to the storage device based on an out place update (OPU) mechanism; and the processing device writes file attribute information corresponding to the target file in the storage device to out of band (OOB) data of the storage device, to restore, after abnormal power-off, access to the target file by using the file attribute information stored in the OOB data. In this way, when writing the target file to the storage device, the processing device may write a file attribute corresponding to the target file to the OOB data, so that after abnormal power-off, access to the target file may be restored by using the file attribute information.