Abstract: Methods, devices, and systems for retrieving information based on cache miss prediction. A prediction that a cache lookup for the information will miss a cache is made based on a history table. The cache lookup for the information is performed based on the request. A main memory fetch for the information is begun before the cache lookup completes, based on the prediction that the cache lookup for the information will miss the cache. In some implementations, the prediction includes comparing a first set of bits stored in the history table with a second set of bits stored in the history table. In some implementations, the prediction includes comparing at least a portion of an address of the request for the information with a set of bits in the history table.

Abstract: Systems and methods related to direct swap caching with noisy neighbor mitigation and dynamic address range assignment are described. A system includes a host operating system (OS), configured to support a first set of tenants associated with a compute node, where the host OS has access to: (1) a first swappable range of memory addresses associated with a near memory and (2) a second swappable range of memory addresses associated with a far memory. The host OS is configured to allocate memory in a granular fashion such that each allocation of memory to a tenant includes memory addresses corresponding to a conflict set having a conflict set size. The conflict set includes a first conflicting region associated with the first swappable range of memory addresses with the near memory and a second conflicting region associated with the second swappable range of memory addresses with the far memory.

Abstract: A method includes receiving, by a memory management unit (MMU) comprising a translation lookaside buffer (TLB) and a configuration register, a request from a processor core to directly modify an entry in the TLB. The method also includes, responsive to the configuration register having a first value, operating the MMU in a software-managed mode by modifying the entry in the TLB according to the request. The method further includes, responsive to the configuration register having a second value, operating the MMU in a hardware-managed mode by denying the request.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for allocating cache resources according to page-level attribute values. In one implementation, the system includes one or more integrated client devices and a cache. Each client device is configured to generate at least a memory request. Each memory request has a respective physical address and a respective page descriptor of a page to which the physical address belongs. The cache is configured to cache memory requests for each of the one or more integrated client devices. The cache comprises a cache memory having multiple ways. The cache is configured to distinguish different memory requests using page-level attributes of respective page descriptors of the memory requests, and to allocate different portions of the cache memory to different respective memory requests.

Abstract: A system includes a memory device and a processing device coupled to the memory device. The processing device can determine a data rate from a first sensor and a data rate from a second sensor. The processing device can write a first set of data received from the first sensor at a first logical block address (LBA) in the memory device. The processing device can write a second set of data received from the second sensor and subsequent to the first set of data at a second LBA in the memory device. The processing device can remap the first LBA and the second LBA to be logically sequential LBAs. The second LBA can be associated with an offset from the first LBA and the offset can correspond to a data rate of the first sensor.

Abstract: An apparatus comprises prefetch circuitry, and a cache having a plurality of entries to store data for access by processing circuitry and blocks of metadata for reference by the prefetch circuitry. The prefetch circuitry can detect one or more access sequences in dependence on training inputs derived from demand accesses processed by the cache in response to memory access operations performed by the processing circuitry. On detecting a given access sequence, this causes an associated given block of metadata providing information indicative of the given access sequence to be stored in a selected entry of the cache. Eviction control circuitry, responsive to a victimisation event, performs an operation to select a victim entry in the cache, the victim entry being selected from one or more candidate victim entries.

Abstract: Systems, apparatus and methods are provided for logical-to-physical (L2P) address translation. A method may comprise receiving a request for a first logical data address (LDA), and calculating a first translation data unit (TDU) index for a first TDU. The first TDU may contain a L2P entry for the first LDA. The method may further comprise searching a cache of lookup directory entries of recently accessed TDUs using the first TDU index, determining that there is a cache miss, generating and storing an outstanding request for the lookup directory entry for the first TDU in a miss buffer, retrieving the lookup directory entry for the first TDU from an in-memory lookup directory, determining that the lookup directory entry for the first TDU is not valid, reserve a TDU space for the first TDU in a memory and generating a load request for the first TDU.

Abstract: Data routing for efficient decompressor use is described. In accordance with the described techniques, a cache controller receives requests from multiple requestors for elements of data stored in a compressed format in a cache. The requests include at least a first request from a first requestor and a second request from a second requestor. A decompression routing system identifies a redundant element of data requested by both the first requestor and the second requestor and causes decompressors to decompress the requested elements of data. The decompression includes performing a single decompression of the redundant element. After the decompression, the decompression routing system routes the decompressed elements to the plurality of requestors, which includes routing the decompressed redundant element to both the first requestor and the second requestor.

Abstract: An exemplary multi-threaded memory management system comprises a memory management unit (MMU) configured with a plurality of physical address (PA) output ports individually dedicated to a respective plurality of threads, wherein the MMU is configured to adjust scheduling of the plurality of threads based on the status of an item requested from a cache. The MMU may be configured to translate a virtual address (VA) input from an individual thread to a PA output on the respective PA output port. The cache may be a translation look-aside buffer. The item requested from the cache may be in transient status when a response is expected or valid status when the response is received. The MMU may signal a thread scheduler to run a thread when a requested item's status becomes valid, permitting stalling individual threads without blocking other threads that continue running using the PA output port dedicated to each thread.

Abstract: A processing system includes a primary processor and a co-processor. The primary processor is couplable to a memory subsystem having at least one memory and operating to execute system software employing memory address translations based on one or more page tables stored in the memory subsystem. The co-processor is likewise couplable to the memory subsystem and operates to perform iterations of a page table walk through one or more page tables maintained for the memory subsystem and to perform one or more page management operations on behalf of the system software based the iterations of the page table walk. The page management operations performed by the co-processor include analytic data aggregation, free list management and page allocation, page migration management, page table error detection, and the like.

Abstract: Technology is disclosed for relocating data in a non-volatile storage system. An integrated memory assembly has a control die and a memory die that contains the memory cells. The control die contains control circuitry that relocates data from one set of physical addresses on the memory die to another set of physical addresses on the memory die. This relocation results in a change of a mapping between logical addresses for the data and the physical addresses for the data. The control circuitry may update an L2P table on the memory die after the relocation to map the logical addresses of the data to the second set of physical addresses. The control die may construct a validity bitmap, which specifies whether data at a physical address is valid or invalid. The foregoing reduces data transfer between the integrated memory assembly and a memory controller, which saves time and power.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for allocating cache resources according to page-level attribute values. In one implementation, the system includes one or more integrated client devices and a cache. Each client device is configured to generate at least a memory request. Each memory request has a respective physical address and a respective page descriptor of a page to which the physical address belongs. The cache is configured to cache memory requests for each of the one or more integrated client devices. The cache comprises a cache memory having multiple ways. The cache is configured to distinguish different memory requests using page-level attributes of respective page descriptors of the memory requests, and to allocate different portions of the cache memory to different respective memory requests.

Abstract: Described apparatuses and methods balance memory-portion accessing. Some memory architectures are designed to accelerate memory accesses using schemes that may be at least partially dependent on memory access requests being distributed roughly equally across multiple memory portions of a memory. Examples of such memory portions include cache sets of cache memories and memory banks of multibank memories. Some code, however, may execute in a manner that concentrates memory accesses in a subset of the total memory portions, which can reduce memory responsiveness in these memory types. To account for such behaviors, described techniques can shuffle memory addresses based on a shuffle map to produce shuffled memory addresses. The shuffle map can be determined based on a count of the occurrences of a reference bit value at bit positions of the memory addresses. Using the shuffled memory address for memory requests can substantially balance the accesses across the memory portions.

Abstract: A neural processing device and a method for using shared page table thereof are provided. The neural processing device including at least one neural processor, a shared memory shared by the at least one neural processor, and a global interconnection configured to exchange data between the at least one neural processor and the shared memory, comprises at least one processing unit each of which included in each of the at least one neural processor and configured to provide logical addresses, a memory management unit configured to receive and translate the logical addresses into physical addresses, and a physical memory accessible by the physical addresses, wherein the memory management unit comprises a shared page table that has translation information between the logical addresses and the physical addresses and is shared by at least one process with each other.

Abstract: A memory sub-system configured to manage programming mode transitions to accommodate a constant size of data transfer between a host system and a memory sub-system. The memory sub-system counts single-page transitions of atomic programming modes performed within a memory sub-system and determines whether or not to allow any two-page transition of atomic programming modes based on whether an odd or even number of the single-page transitions have been counted. When an odd number of the transitions have been counted, no two-page transition is allowed; otherwise, one or more two-page transitions are allowable. A next transition of atomic programming modes is selected based on the determining of whether or not to allow any two-page transitions.

Abstract: To deliver up-to-date, coherent user data to applications upon request, the disclosed technology includes systems and methods for caching data and metadata after it has been synchronously loaded—for future retrieval with a page load time close to zero milliseconds. To provide this experience, data needs to be stored as locally to a user as possible, in the cache on the local device or in an edge cache located geographically nearby, for use in responding to requests. Applications which maintain caches of API results can be notified of their invalidation, and can detect the invalidation, propagate the invalidation to any further client tiers with the appropriate derivative type mapping, and refresh their cached values so that clients need not synchronously make the API requests again—ensuring that the client has access to the most up-to-date copy of data as inexpensively as possible—in terms of bandwidth and latency.

Abstract: A memory system includes a memory device with a memory cell array including a first and second plane and first and second caches. A controller is configured to output status information in response to a status read command. The status information indicating the states of the caches. The controller begins a first process in response to a command addressed to the first plane if the status information indicates the first and second caches are in the ready state, and begins a second process on the second plane according to a second command to the second plane if the status information indicates at least the second cache is in the ready state.

Abstract: A memory device includes a data register operatively coupled to the memory array, a cache operatively coupled to the data register, and an input/output interface operatively coupled to the cache. A controller executes a continuous page read operation to sequentially load pages to the data register and move the pages to the cache, in response to a page read command, executes the cache read operation in response to a cache read command to move data from the cache to the input/output interface, and to stall moving of the data from the cache until a next cache read command, and terminates the continuous page read operation in response to a terminate command.

Abstract: Embodiments of the invention provide systems and methods for managing processing, memory, storage, network, and cloud computing to significantly improve the efficiency and performance of processing nodes. More specifically, embodiments of the present invention are directed to a hardware-based processing node of an object memory fabric.

Abstract: Devices and techniques are disclosed herein to control recovery of a memory device from a reduced power state. A memory controller can include a detection circuit configured to monitor the power supply voltage to an array of memory cells during the reduced power state. Control circuitry an initialization procedure for recovery of the memory device from the reduced power state, based on the state of the detection circuit.