A device includes an arbiter circuit configured to receive a first request for a resource. The first request is associated with a first credit cost. The arbiter circuit is further configured to receive a second request for the resource. The second request is associated with a second credit cost. The arbiter circuit is further configured to select the first request for the resource as an arbitration winner. The arbiter circuit is further configured to decrement a number of available credits associated with the resource by the first credit cost. The arbiter circuit is further configured to, in response to the number of available credits associated with the resource falling to a lower credit threshold, wait until the number of available credits associated with the resource reaches an upper credit threshold to select an additional arbitration winner for the resource.

Interleaved cache controllers with shared metadata are disclosed and described. A memory system may comprise a plurality of cache controllers and a metadata store interconnected by a metadata store fabric. The metadata store receives information from at least one of the plurality of cache controllers, a portion of which is stored as shared distributed metadata. The metadata store provides shared access of the shared distributed metadata hosted to the plurality of cache controllers

A memory having a delayed write-back to the array of data corresponding to a previously opened page allows delays associated with write-back operations to be avoided. After an initial activation opens a first page and the read/write operations for that page are complete, write-back of the open page to the array of memory cells is delayed until after completion of a subsequent activate operation that opens a new page. Techniques to force a write-back in the absence of another activate operation are also disclosed.

This technology relates to a method and apparatus for improving I/O throughput through an interleaving operation for multiple memory dies of a memory system. A memory system may include: multiple memory dies suitable for outputting data of different sizes in response to a read request; and a controller in communication with the multiple memory dies through multiple channels, and suitable for: performing a correlation operation on the read request so that the multiple memory dies interleave and output target data corresponding to the read request through the multiple channels, determining a pending credit using a result of the correlation operation, and reading, from the multiple memory dies, the target data corresponding to the read request and additional data stored in a same storage unit as the target data, based on a type of the target data corresponding to the read request and the pending credit.

A memory system is provided to include a memory device including a plurality of memory blocks and a controller for dynamically changing a size of a write buffer based on whether a current workload is a sequential workload or a mixed workload. The controller includes a workload detecting unit suitable for changing current workload from the sequential workload to the mixed workload based on a read count, or from the mixed workload to the sequential workload based on a write count and a write buffer managing unit suitable for reducing the size of the write buffer when the current workload is changed to the mixed workload.

Described herein are systems, methods, and non-transitory computer readable media for memory address encoding of multi-dimensional data in a manner that optimizes the storage and access of such data in linear data storage. The multi-dimensional data may be spatial-temporal data that includes two or more spatial dimensions and a time dimension. An improved memory architecture is provided that includes an address encoder that takes a multi-dimensional coordinate as input and produces a linear physical memory address. The address encoder encodes the multi-dimensional data such that two multi-dimensional coordinates close to one another in multi-dimensional space are likely to be stored in close proximity to one another in linear data storage. In this manner, the number of main memory accesses, and thus, overall memory access latency is reduced, particularly in connection with real-world applications in which the respective probabilities of moving along any given dimension are very close.

Described herein is a memory architecture that is configured to dynamically determine an address encoding to use to encode multi-dimensional data such as multi-coordinate data in a manner that provides a coordinate bias corresponding to a current memory access pattern. The address encoding may be dynamically generated in response to receiving a memory access request or may be selected from a set of preconfigured address encodings. The dynamically generated or selected address encoding may apply an interleaving technique to bit representations of coordinate values to obtain an encoded memory address. The interleaving technique may interleave a greater number of bits from the bit representation corresponding to the coordinate direction in which a coordinate bias is desired than from bit representations corresponding to other coordinate directions.

A device includes a data path, a first interface configured to receive a first memory access request from a first peripheral device, and a second interface configured to receive a second memory access request from a second peripheral device. The device further includes an arbiter circuit configured to, in a first clock cycle, a pre-arbitration winner between a first memory access request and a second memory access request based on a first number of credits allocated to a first destination device and a second number of credits allocated to a second destination device. The arbiter circuit is further configured to, in a second clock cycle select a final arbitration winner from among the pre-arbitration winner and a subsequent memory access request based on a comparison of a priority of the pre-arbitration winner and a priority of the subsequent memory access request.

A method of dynamic cache configuration includes determining, for a first clustering configuration, whether a current cache miss rate exceeds a miss rate threshold. The first clustering configuration includes a plurality of graphics processing unit (GPU) compute units clustered into a first plurality of compute unit clusters. The method further includes clustering, based on the current cache miss rate exceeding the miss rate threshold, the plurality of GPU compute units into a second clustering configuration having a second plurality of compute unit clusters fewer than the first plurality of compute unit clusters.

This technology relates to a method and apparatus for improving I/O throughput through an interleaving operation for multiple memory dies of a memory system. A memory system may include: multiple memory dies suitable for outputting data of different sizes in response to a read request; and a controller in communication with the multiple memory dies through multiple channels, and suitable for: performing a correlation operation on the read request so that the multiple memory dies interleave and output target data corresponding to the read request through the multiple channels, determining a pending credit using a result of the correlation operation, and reading, from the multiple memory dies, the target data corresponding to the read request and additional data stored in a same storage unit as the target data, based on a type of the target data corresponding to the read request and the pending credit.