Abstract: Techniques for reducing Solid State Device Input/Output latency are disclosed. In some embodiments, the techniques may be realized as a method for reducing Solid State Device Input/Output latency comprising receiving a write request at a Solid State Device, monitoring a plurality flash memory channels of the Solid State Device to identify Input/Output requests, evaluating, using load balancing circuitry, identified Input/Output requests to determine a load of one or more of the plurality of flash memory channels, and assigning a destination flash memory channel out of the plurality of flash memory channels to the write request based on the determined load.

Abstract: An apparatus includes a scheduler module operatively coupled to each memory block from a set of memory blocks via a shared address bus. The scheduler module is configured to receive a group of memory commands from a set of memory controllers. Each memory controller from the set of memory controllers is uniquely associated with a different memory block from the set of memory blocks. The scheduler module is configured to classify each memory command from the group of memory commands into a category based at least in part on memory commands previously sent to the set of memory blocks via the shared address bus. The scheduler module is configured to select an order in which to send each memory command from the group of memory commands to the set of memory blocks via the shared address bus based at least in part on the category of each memory command.

Abstract: A semiconductor device may be provided. The semiconductor device may include an error scrub control circuit and or an active period signal generation circuit. The error scrub control circuit may be configured to generate an error scrub pre-charge signal and an error scrub bank signal for performing an error scrub operation of memory cells included in banks, based on a bank active signal and a row address signal which are generated based on a refresh signal. The active period signal generation circuit may be configured to generate an active period signal from the bank active signal and the error scrub pre-charge signal based on the error scrub bank signal.

Abstract: Memory subsystem refresh management enables commands to access one or more identified banks across different bank groups with a single command. Instead of sending commands identifying a bank or banks in separate bank groups by each bank group individually, the command can cause the memory device to access banks in different bank groups. The command can be a refresh command. The command can be a precharge command.

Abstract: Described herein are embodiments of a process that can be used to balance the allocation of primary memory between different types of information. In some embodiments, the memory allocation is balanced dynamically based on observed I/O patterns. Related system embodiments are also described.

Abstract: A memory apparatus may include a plurality of ranks commonly coupled to an input/output (I/O) terminal. Non-target ranks other than a target rank among the plurality of ranks may be configured to perform an on die termination operation based on a read operation of the target rank.

Abstract: A system includes multiple memories. Access of at least one of the multiple memories uses an interface subsystem that includes a memory controller and a distinct media controller, the memory controller to issue a transaction-level access request. The media controller is associated with at least one memory and produces, in response to the transaction-level access request, at least one command according to a specification of the at least one memory. Data is migrated from a first of the multiple memories to a second of the multiple memories, without the data traversing through a cache memory in the processor during the migrating.

Abstract: Apparatus, systems, and methods to implement polarity based data transfer function for volatile memory power reduction are described. The transfer function take into account certain data values, all zeroes in particular, that are common and transforms them to predetermined values that consume less power and are less common. Similarly, these predetermined values are transformed to the common values.

Abstract: A method for updating a DRAM memory array is disclosed. The method comprises: a) transitioning the DRAM memory array from an idle state to a refresh state in accordance with a command from a memory controller; b) initiating a refresh on the DRAM memory array using DRAM internal control circuitry by activating a row of data into an associated sense amplifier buffer; and c) during the refresh, performing an ERR Correction Code (ECC) scrub operation of selected bits in the activated row of the DRAM memory array.

Abstract: A system and method for decoding command signals that includes a command decoder configured to generate internal control signals to perform an operation based on the command signals and an operating state. The same combination of command signals can request different commands depending on the operating state. A command is selected from a first set of operations according to the command signals when the memory system is in a first operating state and a command is selected from a second set of operations according to the command signals when the memory system is in a second operating state.

Abstract: An impedance calibration circuit is disclosed, which relates to a technology for improving precision of pad resistance. The impedance calibration circuit includes: a first On Die Termination (ODT) circuit selected by a first selection signal, configured to tune its own resistance using a first code signal, and output a first resistance value to an output terminal; and a second ODT circuit selected by a second selection signal, configured to tune its own resistance using a second code signal, and output a second resistance value to the output terminal.

Abstract: A method, a controller, and a system for service flow control in an object-based storage system are disclosed. The method is: receiving, by a controller, a first object IO request; acquiring a processing quantity threshold and a to-be-processed quantity; if the to-be-processed quantity is less than the processing quantity threshold, sending the first object IO request to a storage device client, and updating the to-be-processed quantity; receiving a first response message replied by the storage device client for the first object IO request, where the first response message carries a processing result of the first object IO request; and adjusting the processing quantity threshold according to a received processing result of an object IO request when a preset condition is met. The storage device is not overloaded with object IO requests and can use all resources to effectively, thereby improving performance and a success rate of the object-based storage system.

Abstract: A method for managing a multi-channel memory device includes at least following steps: when the multi-channel memory device is controlled to operate in an M-channel mode, reserving a partial memory space across N memory channels of the multi-channel memory device, where the reserved partial memory space is not used under the M-channel mode, M and N are positive integers, and M is smaller than N; and when the multi-channel memory device is controlled to switch from the M-channel mode to an N-channel mode, accessing data in the reserved partial memory space across the N memory channels used under the N-channel mode. The method for managing a multi-channel memory device can improve switch response time.

Abstract: Methods and apparatus relating to device power management state transition latency advertisement for faster boot time are described. In some embodiments, a storage unit stores a value corresponding to a requisite transition delay period for a first agent to exit from a low power consumption state. The first agent writes the value to the storage unit and a second agent waits for the requisite transition delay period (after the first agent initiates its exit from the low power consumption state) before the second agent attempts to communicate with the first agent via a link. Other embodiments are also disclosed and claimed.

Abstract: Methods and apparatus relating to device power management state transition latency advertisement for faster boot time are described. In some embodiments, a storage unit stores a value corresponding to a requisite transition delay period for a first agent to exit from a low power consumption state. The first agent writes the value to the storage unit and a second agent waits for the requisite transition delay period (after the first agent initiates its exit from the low power consumption state) before the second agent attempts to communicate with the first agent via a link. Other embodiments are also disclosed and claimed.

Abstract: A memory management system for managing a memory and includes a multi-stage memory management unit including control circuitry and cache memory. The cache memory may have a respective translation look-aside buffer for each stage of the multi-stage memory management unit. The control circuitry may be configured to generate a blank data request including a virtual address and information that specifies that data is not to be read from the memory, perform address translations based on the generated blank data request in multiple stages until a physical address is obtained, and discard the blank data request.

Abstract: Disclosed herein are techniques for implementing high-throughput low-latency hybrid memory modules with improved data backup and restore throughput, enhanced non-volatile memory controller (NVC) resource access, and enhanced mode register setting programmability. Embodiments comprise a command replicator to generate sequences of one or more DRAM read and/or write and/or other commands to be executed in response to certain local commands from a non-volatile memory controller (NVC) during data backup and data restore operations. Other embodiments comprise an access engine to enable an NVC in a host control mode to trigger entry into a special mode and issue commands to access a protected register space. Some embodiments comprise a mode register controller to capture and store the data comprising mode register setting commands issued during a host control mode, such that an NVC can program the DRAM mode registers in an NVC control mode.

Abstract: The present disclosure includes apparatuses and methods related to performing comparison operations in memory. An example apparatus can include a first group of memory cells coupled to a first access line and configured to store a plurality of first elements, and a second group of memory cells coupled to a second access line and configured to store a plurality of second elements. The apparatus can include a controller configured to cause the plurality of first elements to be compared with the plurality of second elements by controlling sensing circuitry to perform a number of operations without transferring data via an input/output (I/O) line, and the plurality of first elements and the plurality of second elements can be compared in parallel.

Abstract: A methods and device for accessing virtual machine (VM) data are described. A computing device for accessing virtual machine comprises an access request process module, a data transfer proxy module and a virtual disk. The access request process module receives a data access request sent by a VM and adds the data access request to a request array. The data transfer proxy module obtains the data access request from the request array, maps the obtained data access request to a corresponding virtual storage unit, and maps the virtual storage unit to a corresponding physical storage unit of a distributed storage system. A corresponding data access operation may be performed based on a type of the data access request.

Abstract: Providing for a memory apparatus configured for improved data management for a two-terminal memory array is described herein. By way of example, disclosed embodiments relate to page management and transfer of data between page-sized subsets of a page buffer, and respective pages within one or more memory banks of the two-terminal memory array. The memory apparatus can emulate a larger page size than a physical page buffer utilized by the memory apparatus, to provide compatibility with different page size defaults while lowering current consumption by the page buffer. This can facilitate large or small array operations, taking advantage of higher efficiencies of two-terminal memory devices. In addition, page buffer data management can facilitate interleaved data transfers among multiple banks of memory, facilitating large memory capacities for a disclosed memory apparatus.

Abstract: A method is provided for cache coherence being based on a hybrid approach relying on hardware-and software-implemented functionalities. In case a processor core is requested to perform a write operation on a memory line missed in the local cache of said core, a hardware-implemented coherence directory ensures that said processor core becomes assigned exclusive write permissions to indicate that the memory line in said local cache is up-to-date after said write. In case the processor core is requested to perform a read operation on a memory line missed in the local cache of said processor core, the coherence directory updates the coherence directory to indicate that none of the processor cores of the system has exclusive write permission on the memory line and relies on software executed on said processor core to ensure that the cached memory line is up-to-date before performing the read operation.

Abstract: A write method and a write apparatus for a storage device, where the write method includes: acquiring n numerical values that need to be written; determining n bits corresponding to the n numerical values, and information about a stuck-at fault included in the n bits; grouping the n bits into B groups of bits, so that the B groups of bits meet a grouping condition; and correspondingly writing the n numerical values according to information about a stuck-at fault included in each group of bits in the B groups of bits and a numerical value that needs to be written and that is corresponding to the information about the stuck-at fault included in each group of bits in the B groups of bits.

Abstract: The disclosed embodiments relate to a memory system that generates a multiplied timing signal from a reference timing signal. During operation, the system receives a reference timing signal. Next, the system produces a multiplied timing signal from the reference timing signal by generating a burst comprising multiple timing events for each timing event in the reference timing signal, wherein consecutive timing events in each burst of timing events are separated by a bit time. Then, as the reference clock frequency changes, the interval between bursts of timing events changes while the bit time remains substantially constant.

Abstract: An apparatus includes a detector circuit and a data buffer. The detector circuit may be configured to (i) identify a start of a command sequence associated with a directed access to a memory system and (ii) generate a control signal indicating a non-consecutive clock associated with the start of said command sequence. The data buffer circuit may be configured to initialize a condition of a receiver circuit in response to the control signal prior to reception of a first data bit associated with the command sequence.

Abstract: Provided are a method and apparatus for method and apparatus for encoding registers in a memory module. A mode register command is sent to the memory module over a bus, initialization of the memory module before the bus to the memory module is trained for bus operations, to program one of a plurality of mode registers in the memory module, wherein the mode register command indicates one of the mode registers and includes data for the indicated mode register.

Abstract: The method for maintaining a storage mapping table is introduced. After a total number of logical blocks, which exceeds a specified number, have been programmed into a storage unit, an access interface is directed to program a corresponding group of a storage mapping table of a DRAM (Dynamic Random Access Memory) into a first block of the storage unit according to a group number of an unsaved group queue. A group mapping table of the DRAM is updated to indicate that the latest data of the group of the storage mapping table is stored in which location in the storage unit. The group number is removed from the unsaved group queue.

Abstract: A multiple memory rank selection method and system assigns, based at least in part on decoding an assignment signal in a second command/address signal, a first terminal of a memory device to receive a first command/address signal and a second terminal of the memory device to receive the second command/address signal or assigns the first terminal of the memory device to receive the second command/address signal and the second terminal of the memory device to receive the first command/address signal. The multiple memory selection method and system decodes a selection signal encoded in the first command/address signal and enables the memory device based at least in part on the assignment signal and the selection signal.

Abstract: There is provided a storage apparatus connected to a storage device that stores data and a cache device used as a temporary storage area of the data. The storage apparatus includes: a memory configured to store location information for associating a storage location of data stored in the storage device with content information relating to content of the data, and cache information including the content information corresponding to data stored in the cache device; and a processor configured to identify, upon receiving a read request specifying the storage location and requesting to read data, the content information corresponding to the storage location specified in the read request and, when the content information is included in the cache information, read the data from the cache device and return the data as a response.

Abstract: A semiconductor memory device includes a first memory die having a first termination resistor for an on-die termination and a second memory die having a second termination resistor for an on-die termination and formed on the first memory die. Each of the first and second memory dies has a center pad type and operates based on a multi-rank structure. When the first memory die is accessed, the second termination resistor is connected to the second memory die, and when the second memory die is accessed, the first termination resistor is connected to the first memory die.

Abstract: A semiconductor memory device includes: a memory cell array including a plurality of word lines; a word line driving unit suitable for activating a first word line among the plurality of word lines corresponding to an input address signal; an activation time detection unit suitable for enabling a detection signal by detecting an activation time of the first word line; an address latch unit suitable for latching an address information for a second word line corresponding to the first word line in response to the detection signal; and an address output unit suitable for providing the word line driving unit with the latched address information for the second word line during a refresh operation.

Abstract: A method for updating a DRAM memory array is disclosed. The method comprises: a) receiving a command from a memory controller to initiate an active cycle for activating a memory row in a DRAM memory array; b) performing an Error Correction Code (ECC) scrub on the memory row prior to reading data from the memory row into sense amplifiers in the DRAM memory array in accordance with the command to activate; c) activating the memory row; and d) writing corrected data following the ECC scrub back into memory from the sense amplifiers during a pre-charge cycle of the DRAM memory array.

Abstract: Methods and apparatuses are disclosed for requesting ready status information from a memory. One example apparatus includes a memory and a host coupled to the memory. The host is configured to provide a plurality of memory access requests to the memory, to request ready status information regarding whether the memory is ready to execute a memory access request of the plurality of memory access requests, and to request execution of the memory access request responsive to the ready status information.

Abstract: This invention describes an apparatus, computer architecture, memory structure, memory control, and cache memory operation method for multi-core processor. A logic core shares requests when faced with immediate cache memory units having low yield or deadly performance. The core mounts (multiple) cache unit(s) that might already be in use by other logic cores. Selected cache memory units serve multiple logic cores with the same contents. The shared cache memory unit(s) serves all the mounting cores with cache search, hit, miss, and write back functions. The method recovers a logic core whose cache memory block is not operational by sharing cache memory blocks which might already engage other logic cores. The method is used to improve reliability and performance of the remaining system.

Abstract: Disclosed aspects include managing the access of flash memory by a computer system. A physical memory address space which includes a flash memory portion is established. The flash memory portion may correspond to an input/output memory range. An access request may be detected with respect to the physical memory address space. Using a load-store technique to process the access request, the flash memory portion of the physical memory address space may be accessed.

Abstract: An apparatus includes a scheduler module operatively coupled to each memory block from a set of memory blocks via a shared address bus. The scheduler module is configured to receive a group of memory commands from a set of memory controllers. Each memory controller from the set of memory controllers is uniquely associated with a different memory block from the set of memory blocks. The scheduler module is configured to classify each memory command from the group of memory commands into a category based at least in part on memory commands previously sent to the set of memory blocks via the shared address bus. The scheduler module is configured to select an order in which to send each memory command from the group of memory commands to the set of memory blocks via the shared address bus based at least in part on the category of each memory command.

Abstract: An apparatus and method are provided for controlling access to a memory device. The apparatus has a pending access requests storage that is used to store access requests waiting to be issued to the memory device, and memory access control circuitry is then used to issue to the memory device access requests selected from the pending access requests storage. Access requests are received at an interface of the apparatus from at least one requesting device, and access request evaluation circuitry within the apparatus is arranged to apply criteria to determine, for a current access request, whether to accept that current access request or reject that current access request. The criteria applied takes account of at least one access timing characteristic of the memory device. The access request evaluation circuitry is responsive to determining that the current access request is to be accepted, to cause that current access request to be stored in the pending access requests storage.

Abstract: A stacked memory with interface providing offset interconnects. An embodiment of memory device includes a system element and a memory stack coupled with the system element, the memory stack including one or more memory die layers. Each memory die layer includes first face and a second face, the second face of each memory die layer including an interface for coupling data interface pins of the memory die layer with data interface pins of a first face of a coupled element. The interface of each memory die layer includes connections that provide an offset between each of the data interface pins of the memory die layer and a corresponding data interface pin of the data interface pins of the coupled element.

Abstract: A buffer manager is generated by executing a script with respect to a buffer architecture template and a configuration file specifying parameters for the buffer such as, for example, number of memory banks, width of memory banks, depth of memory banks, and client bridge FIFO depth. The script converts the buffer architecture template into a hardware description language (HDL) description of a buffer manager having the parameters. Client bridges accumulate requests for memory banks in FIFO that is provided to a buffer manager upon the client bridge being granted arbitration. Accesses of memory banks may be performed one at a time in consecutive clock cycles in a pipelined manner. Client bridges and the buffer manager may operate in different clock domains. The clock frequency of the buffer manager may be increased or decreased according to requests from client devices.

Abstract: Examples of the present disclosure provide apparatuses and methods related to performing division operations in memory. An example apparatus might include a first group of memory cells coupled to a first access line and configured to store a dividend element. An example apparatus might include a second group of memory cells coupled to a second access line and configured to store a divisor element. An example apparatus might also include a controller configured to cause the dividend element to be divided by the divisor element by controlling sensing circuitry to perform a number of operations without transferring data via an input/output (I/O) line.

Abstract: Techniques for managing content stored on persistent memory modules so as to ensure that the content can be accessed from the persistent memory modules in the correct order are provided. In one aspect, a method for managing content stored on persistent memory modules is provided. The method includes the step of: on each of the persistent memory modules, storing configuration data relating to an order in which the content is stored on the persistent memory modules. A method for managing content stored on persistent memory modules in a system is also provided.

Abstract: A method for operating a memory system includes receiving thermal data indicating a temperature at addresses in a memory array, and a write request associate with data. An address of the write request is decoded. It is determined whether a temperature at the address of the write request is above a threshold temperature. The data is sent to a short latency write queue responsive to determining that the temperature is not above the threshold temperature.

Abstract: A memory device includes a memory component that store data and a processor. The processor may generate one or more data packets associated with the memory component. Each data packet may include a transaction type field that includes data indicative of a first size of a payload of the respective data packet and a second size of an error control code in the respective data packet. Each packet may also have a payload field that includes the payload and an error control code field that includes the error control code. The processor may transmit the data packets to a requesting component, such that the requesting component identifies the payload field and the error control field of each data packet based on the data of the transaction type field in each data packet.

Abstract: During a command/address calibration mode, a memory controller may transmit multiple cycles of test patterns as signals to a memory device. Each cycle of test pattern signals may be transmitted at an adjusted relative phase with respect to a clock also transmitted to the memory device. The memory device may input the test pattern signals at a timing determined by the clock, such as rising and/or falling edges of the clock. The test pattern as input by the memory device may be sent to the memory controller to determine if the test pattern was successfully transmitted to the memory device during the cycle. Multiple cycles of test pattern transmissions are evaluated to determine a relative phase of command/address signals with respect to the clock for transmission during operation of the system.

Abstract: During runtime of a system, a memory controller is caused to relinquish control of a memory module that includes a volatile memory and a non-volatile memory. After the triggering, an indication is activated to the memory module, the indication causing a backup operation in the memory module, the backup operation being controlled by an internal controller in the memory module, and the backup operation involving a transfer of data from the volatile memory to the non-volatile memory in the memory module.

Abstract: A semiconductor memory device includes a first memory die having a first termination resistor for an on-die termination and a second memory die having a second termination resistor for an on-die termination and formed on the first memory die. Each of the first and second memory dies has a center pad type and operates based on a multi-rank structure. When the first memory die is accessed, the second termination resistor is connected to the second memory die, and when the second memory die is accessed, the first termination resistor is connected to the first memory die.

Abstract: Systems and methods are disclosed for managing program code in a computing device, such as an embedded system. In a computing device, a non-volatile flash memory stores program code comprising initialization code designed to be executed a single time during device booting, main program code separate from the initialization code, and main code loader code designed to direct a controller of the computing device to load the main program code from the flash memory to the RAM when executed by the controller. The controller loads the initialization code and the main code loader code to first and second adjacent portions of the RAM, respectively, and loads the main program code, separately from the initialization code, to a third portion of the RAM at least partially overlapping the first portion but not overlapping the second portion, thereby at least partially overwriting the initialization code.

Abstract: An integrated circuit device comprises a first data processing element having a first data interface configured to synchronously communicate data according to a first clock signal at a first clock speed, and a second data interface configured to synchronously communicate data at a second clock speed lower than the first clock speed; and a second data processing element configured to operate in response to a second clock signal at the second clock speed and to synchronously communicate data with the first data processing element via the second data interface according to the second clock speed; the first data processing element being configured to derive, from a source clock signal, the first clock signal and the second clock signal; and the first data processing element and the second data processing element each comprising a clock signal interface by which the second clock signal is provided by the first data processing element to the second data processing element.

Abstract: Memory subsystem error management enables dynamically changing lockstep partnerships. A memory subsystem has a lockstep partnership relationship between a first memory portion and a second memory portion to spread error correction over the pair of memory resources. The lockstep partnership can be preconfigured. In response to detecting a hard error in the lockstep partnership, the memory subsystem can cancel or reverse the lockstep partnership between the first memory portion and the second memory portion and create or set a new lockstep partnership. The detected error can be a second hard error in the lockstep partnership. The memory subsystem can create new lockstep partnerships between the first memory portion and a third memory portion as lockstep partners and between the second memory portion and a fourth memory portion as lockstep partners. The memory subsystem can also be configured to change the granularity of the lockstep partnership when changing partnerships.

Abstract: A table may include first and second row addresses that are adjacent an activated row address. A first counter of the first row address may be incremented if the activated row address is not included in the table. A second counter of the second row address may also be incremented if the activated row address is not included in the table. The first row address may be refreshed if the first counter exceeds a counter threshold. The second row address may be refreshed if the second counter exceeds the counter threshold.

Abstract: A data storage apparatus and method of storing data in a data storage apparatus are provided, where the data storage apparatus comprises multiple banks for storing data. The multiple banks form multiple bank groups, wherein each bank group comprising more than one bank. A first data item of a received data block is stored at a selected storage location in a selected bank and a subsequent data item of the data block is stored to a further storage location in a different bank according to a sequence of banks. The sequence of banks firstly comprises the selected bank followed by all other banks in the bank group of the selected bank. Moreover the sequence of banks respects a hierarchical pattern, wherein a finer granularity of the hierarchical pattern comprises all banks in a given bank group, and a coarser granularity of the hierarchical pattern comprises the given bank group followed by a different bank group to the given bank group. Access to the data storage apparatus is thereby improved.