A memory system of an artificial neural network (ANN) includes a processor configured to process an ANN model; and an ANN memory controller configured to control a rearrangement of data of the ANN model stored in a memory and to operate the data of the ANN model stored in the memory in a read-burst mode based on ANN data locality information of the ANN model. The ANN memory controller may receive pre-generated ANN data locality information, or the processor may generate a plurality of data access requests sequentially so that the ANN memory controller may generate the ANN data locality information by monitoring the plurality of data access requests. The ANN memory controller prepares, based on an artificial neural network data locality, data before receiving a request from the processor in order to reduce a delay in the data supply of the memory to the processor.

A memory system of an artificial neural network (ANN) includes a processor configured to process an ANN model; and an ANN memory controller configured to control a rearrangement of data of the ANN model stored in a memory and to operate the data of the ANN model stored in the memory in a read-burst mode based on ANN data locality information of the ANN model. The ANN memory controller may receive pre-generated ANN data locality information, or the processor may generate a plurality of data access requests sequentially so that the ANN memory controller may generate the ANN data locality information by monitoring the plurality of data access requests. The ANN memory controller prepares, based on an artificial neural network data locality, data before receiving a request from the processor in order to reduce a delay in the data supply of the memory to the processor.

Dynamically coalescing atomic memory operations for memory-local computing is disclosed. In an embodiment, it is determined whether a first atomic memory access and a second atomic memory access are candidates for coalescing. In response to a triggering event, the atomic memory accesses that are candidates for coalescing are coalesced in a cache prior to requesting memory-local processing by a memory-local compute unit. The atomic memory accesses may be coalesced in the same cache line or atomic memory accesses in different cache lines may be coalesced using a multicast memory-local processing command.

The present disclosure includes apparatuses and methods related to a memory system with cache line data. An example apparatus can store data in a number of cache lines in the cache, wherein each of the number of lines includes a number of chunks of data that are individually accessible.

The present disclosure includes apparatuses and methods related to a memory system with cache line data. An example apparatus can store data in a number of cache lines in the cache, wherein each of the number of lines includes a number of chunks of data that are individually accessible.

Operational information in a storage system is collected regarding storage media storage tiers, devices, drives, tracks on drives, and logical storage layers, to determine an estimated amount of time it will take to write data from cache to the intended drive when a new write operation arrives at the storage system. This information is then used to decide which type of cache is most optimal to store the data for the write operation, based on the estimated amount of time it will take to write data out from the cache. By allocating cache slots from a faster cache to write operations that are expected to quickly be written out to memory, and allocating cache slots from the slower cache to write operations that are expected to take more time to be written out to memory, it is possible to increase the availability of the cache slots in the faster cache.

Operational information in a storage system is collected regarding storage media storage tiers, devices, drives, tracks on drives, and logical storage layers, to determine an estimated amount of time it will take to write data from cache to the intended drive when a new write operation arrives at the storage system. This information is then used to decide which type of cache is most optimal to store the data for the write operation, based on the estimated amount of time it will take to write data out from the cache. By allocating cache slots from a faster cache to write operations that are expected to quickly be written out to memory, and allocating cache slots from the slower cache to write operations that are expected to take more time to be written out to memory, it is possible to increase the availability of the cache slots in the faster cache.

An apparatus includes a central processing unit (CPU) core and a cache subsystem coupled to the CPU core. The cache subsystem includes a first memory, a second memory, and a controller coupled to the first and second memories. The controller is configured to execute a sequence of scrubbing transactions on the first memory and execute a functional transaction on the second memory. One of the scrubbing transactions and the functional transaction are executed concurrently.

An apparatus includes a central processing unit (CPU) core and a cache subsystem coupled to the CPU core. The cache subsystem includes a first memory, a second memory, and a controller coupled to the first and second memories. The controller is configured to execute a sequence of scrubbing transactions on the first memory and execute a functional transaction on the second memory. One of the scrubbing transactions and the functional transaction are executed concurrently.

An apparatus includes a central processing unit (CPU) core and a cache subsystem coupled to the CPU core. The cache subsystem includes a memory configured to store a line of data and an error correcting code (ECC) syndrome associated with the line of data, where the ECC syndrome is calculated based on the line of data and the ECC syndrome is a first type ECC. The cache subsystem also includes a controller configured to, in response to a request from a master configured to implement a second type ECC, the request being directed to the line of data, transform the first type ECC syndrome for the line of data to a second type ECC syndrome send a response to the master. The response includes the line of data and the second type ECC syndrome associated with the line of data.