Disclosed are a solid state drive and a write operation method. The solid state drive comprises: a controller, receiving write data from outside and comprising a first cache unit for storing the write data; a Flash memory, receiving the write data sent by the first cache unit according to a first instruction of the controller; a second cache unit, storing the write data from the first cache unit as backup data, and sending the backup data to the Flash memory according to a second instruction of the controller. The second instruction is obtained after the write data fails to be written into the Flash memory under the first instruction, so that the backup data can continue to be called if write operation fails. By combining advantages of the first and second cache units, efficiency and quality of write operations are improved and bandwidth requirements are lowered.

Various embodiments provide for one or more processor instructions and memory instructions that enable a memory sub-system to copy, move, or swap data across (e.g., between) different memory tiers of the memory sub-system, where each of the memory tiers is associated with different memory locations (e.g., different physical memory locations) on one or more memory devices of the memory sub-system.

A technique performs data reduction on host data of a write request during ingest under certain circumstances. Therein, raw host data of a write request is placed from the host into a data cache. Further, a data reducing ingest operation is performed that reduces the raw host data from the data cache into reduced host data (e.g., via deduplication, compression, combinations thereof, etc.). After completion of the data reducing ingest operation, performing a late-binding operation is performed that updates a mapper with ability to access the reduced host data from secondary storage. Such ingest-time data reduction may be enabled/disabled (e.g., turned on or off) per input/output (I/O) operation (e.g., used only for relatively large asynchronous I/O operations) and/or activated in situations in which the ingest bandwidth is becoming a bottleneck.

Data can be prefetched in a distributed storage system. For example, a computing device can receive a message with metadata associated with at least one request for an input/output operation from a message queue. The computing device can determine, based on the message from the message queue, an additional IO operation predicted to be requested by a client subsequent to the at least one request for the IO operation. The computing device can send a notification to a storage node of a plurality of storage nodes associated with the additional IO operation for prefetching data of the additional IO operation prior to the client requesting the additional IO operation.

The disclosed embodiments disclose techniques for managing a distributed cache in a cloud-based distributed computing environment (CBDCE). During operation, an instance of a data processing layer service (DPL) receives a data request from a client that specifies an address and an operation for a target data block. The DPL instance uses these to determine a first cache instance of the distributed cache that is assigned to cache a metadata entry that links the address with a data block fingerprint for the target data block. The DPL instance then uses the data block fingerprint and the cache mapping to determine a second cache instance that is assigned to store the target data block, and then accesses the second cache instance to complete the operation for the target data block.

Aspects of a storage device are provided that allow a controller to leverage cache to minimize occurrence of HMB address overlaps between different HMB requests. The storage device may include a cache and a controller coupled to the cache. The controller may store in the cache, in response to a HMB read request, first data from a HMB at a first HMB address. The controller may also store in the cache, in response to an HMB write request, second data from the HMB at a second HMB address. The controller may refrain from processing subsequent HMB requests in response to an overlap of the first HMB address with an address range including the second HMB address, and the controller may resume processing the subsequent HMB requests after the first data is stored. As a result, turnaround time delays for HMB requests may be reduced and performance may be improved.

Disclosed herein are methods, systems, and processes to provide coherency across disjoint caches in clustered environments. It is determined whether a data object is owned by an owner node, where the owner node is one of multiple nodes of a cluster. If the owner node for the data object is identified by the determining, a request is sent to the owner node for the data object. However, if the owner node for the data object is not identified by the determining, selects a node in the cluster is selected as the owner node, and the request for the data object is sent to the owner node.

A data storage device includes a non-volatile memory (NVM) device and a controller coupled to the NVM device. The controller is configured to create a bad block table that tracks bad blocks of the NVM device, send the bad block table to a host memory location, and check the bad block table to determine whether a block to be read or written to is bad. The controller is further configured to request information on a bad block from the bad block table located in the host memory location, determine that the requested information is not available from the host memory location, and retrieve the requested information from a location separate from the host memory location. A sum of the times to generate a request to check the flat relink table, execute the request, and retrieve the requested information is less than a time to process a host command.

A system and method for network traffic management between multiple nodes are described. A computing system includes multiple nodes connected to one another. When a home node determines a number of nodes requesting read access for a given data block assigned to the home node exceeds a threshold and a copy of the given data block is already stored at a first node of the multiple nodes in the system, the home node sends a command to the first node. The command directs the first node to forward a copy of the given data block to the home node. The home node then maintains a copy of the given data block and forwards copies of the given data block to other requesting nodes until the home node detects a write request or a lock release request for the given data block.

A method for constructing a persistent memory index in a non-uniform memory access architecture includes: maintaining partial persistent views in a persistent memory and maintaining a global volatile view in a DRAM; an underlying persistent memory index processing a request in a foreground thread when cold data is accessed; when hot data is accessed, reading a key-value pair for a piece of hot data in the global volatile view in response to a query operation carried in the request, and in response to an insert/update/delete operation carried in the request, updating a local partial persistent view and the global volatile view; and in response to a hotspot migration, a background thread generating new partial persistent views and a new global volatile view, and recycling the partial persistent views and the global volatile view for old hot data into the underlying persistent memory index.