Abstract: The wear-leveling techniques include discovering a persistent state of one or more memory devices, or building and caching persistent state parameters for each logical unit of a given memory device if a persistent state is not discovered for a given memory device. The techniques may also include processing memory access commands utilizing the cached persistent state parameters. When processing memory access commands, the logical block address and length parameter of a logical address of a command may be translated to a plurality of physical addresses for accessing one or more memory devices, each physical address includes a device address, a logical unit address, a block address, and a page address, wherein the block address includes one or more interleaved address bits.

Abstract: Systems and methods are disclosed to improve the performance of a memory system by freeing up physical memory areas that correspond to logical block address ranges that have repeated data patterns. A controller detects data patterns in incoming data. When a data pattern is detected, the data is not written to non-volatile storage area. Rather, the logical block address range of the data is marked in a data structure as having pattern data. The pattern may also be recorded in the data structure as a pattern descriptor. Because the data having the data pattern is not written to the non-volatile storage area, the freed up corresponding physical memory area may be utilized by the memory system for other purposes, thereby improving the overall performance and endurance of the memory system.

Abstract: A memory system is provided. The system includes a volatile memory, a refresh counter configured to monitor a number of advanced refreshes performed in the volatile memory, and a controller configured to check the refresh counter to determine whether a regularly scheduled refresh can be skipped in response to detecting a request for the regularly scheduled refresh.

Abstract: Virtual chip enable techniques perform memory access operations on virtual chip enables rather than physical chip enables. Each virtual chip enable is a construct that includes attributes that correspond to a unique physical or logical memory device.

Abstract: A network device allocates a particular number of memory blocks in a ternary content-addressable memory (TCAM) of the network device to each database of multiple databases, and creates a list of additional memory blocks in an external TCAM of the network device. The network device also receives, by the external TCAM, a request for an additional memory block to provide one or more rules from one of the multiple databases, and allocates, by the external TCAM and to the requesting database, an additional memory block from the list of additional memory blocks.

Abstract: A computer system includes at least one processor, multiple memory modules embodying a main memory, a communications medium for communicating data between the at least one processor and main memory, and memory access control logic which controls the routing of data and access to memory. The communications medium and memory access control logic are designed to accommodate a heterogenous collection of main memory configurations, in which at least one physical parameter is variable for different configurations. The bits of the memory address are mapped to actual memory locations by assigning fixed bit positions to the most critical physical parameters across multiple different module types, and assigning remaining non-contiguous bit positions to less critical physical parameters. In the preferred embodiment, the computer system employs a distributed memory architecture.

Abstract: A system monitors and dynamically changes memory mapping in a runtime of a computing system. The computing system has various memory resources, and multiple possible mappings that indicate how data is to be stored in and subsequently accessed from the memory resources. The performance of each memory mapping may be different under different runtime or load conditions of the computing device. A memory controller can monitor runtime performance of the current memory mapping and dynamically change memory mappings at runtime based on monitored or observed performance of the memory mappings. The performance monitoring can be modified for any of a number of different granularities possible within the system, from the byte level to memory channel.

Abstract: Embodiments of the invention generally pertain to memory devices and more specifically to reducing the write amplification of memory devices without increasing cache requirements. Embodiments of the present invention may be represented as a modified B+ tree in that said tree comprises a multi-level tree in which all data items are stored in the leaf nodes of the tree. Each non-leaf node in the tree will reference a large number of nodes in the next level down from the tree. Modified B+ trees described herein may be represented as data structures used to map memory device page addresses. The entire modified B+ tree used to map said pages may be stored on the same memory device requiring limited amounts of cache. These embodiments may be utilized by low cost controllers that require good sequential read and write performance without large amounts of cache.

Abstract: A single interconnect is provided between a first processor and a second processor, such that the first processor may access a common memory through the second processor while the second processor can be mostly powered off. The first processor accesses the memory through a memory controller using a standard dynamic random access memory (DRAM) bus protocol. Instead of the memory controller directly connecting to the memory, the access path is through the second processor to the memory. Additionally, a bidirectional communication protocol bus is mapped to the existing DRAM bus signals. When both the first processor and the second processor are active, the bus protocol between the processors switches from the DRAM protocol to the bidirectional communication protocol. This enables the necessary chip-to-chip transaction semantics without requiring the additional cost burden of a dedicated interface for the bidirectional communication protocol.

Abstract: Even if each processor core uses the same logical address, a processing-target program corresponding to each processor core can be selected. A logical address of multiple address mapping tables is set to a same logical address in correspondence with an embedded OS program or a RAID management program, and a physical address is set to a different physical address in correspondence with the actual storage destination of an embedded OS program or a RAID management program. Each processor core, on start-up, uses a self address mapping table to execute address mapping processing with each processor core based on the same logical address, selects an embedded OS program or a RAID management program according to the physical address obtained in the address mapping processing, and executes processing according to the selected program.

Abstract: System, method, computer program product embodiments and combinations and sub-combinations thereof for managing data storage as an in-memory database in a database management system (DBMS) are provided. In an embodiment, a specialized database type is provided as a parameter of a native DBMS command. A database hosted entirely in-memory of the DBMS is formed when the specialized database type is specified.

Abstract: A method for data distribution, including distributing logical addresses among an initial set of devices so as provide balanced access, and transferring the data to the devices in accordance with the logical addresses. If a device is added to the initial set, forming an extended set, the logical addresses are redistributed among the extended set so as to cause some logical addresses to be transferred from the devices in the initial set to the additional device. There is substantially no transfer of the logical addresses among the initial set. If a surplus device is removed from the initial set, forming a depleted set, the logical addresses of the surplus device are redistributed among the depleted set. There is substantially no transfer of the logical addresses among the depleted set. In both cases the balanced access is maintained.

Abstract: Provided are an apparatus and a method of reducing memory access conflict. An apparatus for reducing memory access conflict when a plurality of data processing elements perform simultaneous access to a memory including a plurality of pages, each of which includes a plurality of subpages, the apparatus comprising: an access arbiter mapping a subpage division address corresponding to least significant bits of a memory access address received from each of the data processing elements to another address having a same number of bits as the subpage division address in order for data to be output from each of the subpages in a corresponding page at a time of the simultaneous access; and a selector, prepared for each of the pages, selecting to output one of the data output from the subpages using the mapped results.

Abstract: A system including a memory system and a memory controller is connected to a host system. The memory system has at least one memory device storing data. The controller translates the requests from the host system to one or more separatable commands interpretable by the at least one memory device. Each command has a modular structure including an address identifier for one of the at least one memory devices and a command identifier representing an operation to be performed by the one of the at least one memory devices. The at least one memory device and the controller are in a series-connection configuration for communication such that only one memory device is in communication with the controller for input into the memory system. The memory system can include a plurality of memory devices connected to a common bus.

Abstract: A method for updating, in the background, data stored in physical memories without affecting the current operations performed by the microprocessor. When the update is completely terminated, the application switches from an old version to a new version. This switching occurs by a reconfiguration of the page table during which a first sub-tree structure of pointers accessing the old version of data stored in memories is replaced by a second sub-tree structure of pointers thus allowing access to the new version of data. This update method prevents incoherent transitory states of the system as the latter works with the previous data version until the installation of the new version becomes usable. In the case of an interruption to the update process, the application can always reinitialize the update since the old version of data can be reactivated by returning to the previous configuration of the page table.

Abstract: Memory apparatus and methods utilizing multiple bit lanes may redirect one or more signals on the bit lanes. A memory agent may include a redrive circuit having a plurality of bit lanes, a memory device or interface, and a fail-over circuit coupled between the plurality of bit lanes and the memory device or interface.

Abstract: Deduplication of data using a low-latency random read memory (LLRRM) is described herein. Upon receiving a block, if a matching block stored on a disk device is found, the received block is deduplicated by producing an index to the address location of the matching block. In some embodiments, a matching block having a predetermined threshold number of associated indexes that reference the matching block is transferred to LLRRM, the threshold number being one or greater. Associated indexes may be modified to reflect the new address location in LLRRM. Deduplication may be performed using a mapping mechanism containing mappings of deduplicated blocks to matching blocks, the mappings being used for performing read requests. Deduplication described herein may reduce read latency as LLRRM has relatively low latency in performing random read requests relative to disk devices.

Abstract: A method and system for refreshing DRAM having a plurality of banks, each of the banks including a plurality of rows includes dividing all banks in DRAM into a plurality of groups of banks, each of the groups having n banks, wherein n is an integer greater than or equal to 1. A threshold of available retention time for each group of banks is determined. Each row of banks in each group of banks is refreshed. Refreshing one row of a bank in one group of banks includes determining whether a refresh operation for the row of the bank conflicts with an access operation for the bank where the row of the bank is located. If there is a conflict, then it is determined whether to perform the refresh operation or the access operation for the current row of the bank. If it is determined to perform the access operation, the access operation is continued. If it is determined to perform the recess operation, the current row of the bank is refreshed. DRAM access performance is improved.

Abstract: A device may include a line card and a control module. The line card may include a memory that stores a local routing table. The line card may request a routing entry from a routing table, receive the routing entry, insert the routing entry in the local routing table, and age out stale routing entries from the local routing table. The control module may include the routing table. The control module may distribute the routing entry in the routing table to the line card.

Abstract: Memory access arbitration allowing a shared memory to be used both as a memory for a processor and as a buffer for data flows, including an arbiter unit that makes assignment for access requests to the memory sequentially and transfers blocks of data in one round-robin cycle according to bandwidths required for the data transfers, sets priorities for the transfer blocks so that the bandwidths required for the data transfers are met by alternate transfer of the transfer blocks, and executes an access from the processor with an upper limit set for the number of access times from the processor to the memory in one round-robin cycle so that the access from the processor with the highest priority and with a predetermined transfer length exerts less effect on bandwidths for data flow transfers in predetermined intervals between the transfer blocks.

Abstract: A memory controller includes a memory capacity setting circuit and an address selecting circuit. The memory capacity setting circuit is configured to set a valid memory capacity of a memory device based on a defective cell information signal, and generate a valid memory capacity signal based on the valid memory capacity. The address selecting circuit is configured to disable an address signal corresponding to a memory block having a defective cell, and generate a selection address signal based on the valid memory capacity signal and the disabled address signal. A non-defective cell in a memory cell array is activated based on the selection address signal and a command signal.

Abstract: An embedded device is hibernated by storing state data of the embedded device to a non-volatile data storage medium, and powering off the embedded device. The embedded device is later woken up in response to the detection of a wakeup event from a wakeup source. The state data stored in the RAM of the embedded device comprises data in one or more registers of a Central Processing Unit (CPU) of the embedded device, one or more registers of a system-on-chip (SOC) of the embedded device, and the system and applications code and data. Waking the embedded device comprises loading, from the non-volatile data storage medium, initial memory sections that are used to run a kernel of the embedded device. State data that is stored in the RAM of a system may be compressed by dividing the RAM into a plurality of sections and independently choosing, for each section in the plurality of sections, a corresponding compression arithmetic.

Abstract: A memory-array decoder operably coupled to a memory array comprising a sequence of rows and receiving as input a plurality of address bits whereby these address bits are transformed by transforming logic. This transforming logic may include adders. Transforming logic may alternately include comparators or exclusive-or circuits. Transforming logic comprising adders may include overflow carry bits that are discarded, ignored, or otherwise not used or the overflow logic may be omitted altogether.

Abstract: The present disclosure describes various techniques resolving block boundary issues and reconstructing logical blocks in a block access storage device when there are resulting mismatches between logical and physical block sizes or alignments, such that logical blocks span multiple physical block boundaries in irregular ways. In one example, a method comprises the following features: receiving logical block addresses that are associated with a sequence of logical blocks; and locating a first portion of a logical block within a first physical block that is stored in a block access storage device based upon a logical block address of the logical block, wherein the logical block is part of the sequence of logical blocks, and wherein at least two logical blocks within the sequence of logical blocks have different sizes.

Abstract: A processor 6 is provided with a plurality of hardware resources, such as performance monitors 12 and context pointers 18. Boundary indicating circuitry 14, 20 stores a boundary value which is programmable and which indicates a boundary position dividing the hardware resources into a first portion and a second portion. Resource control circuitry 16, 22 controls access to the hardware resources such that when program execution circuitry 8 is executing a first program it is responsive to a query as to how many off said plurality of hardware resources are present to return a first value whereas when the program execution circuitry is executing a second program it responds to such a query by returning a value corresponding to those hardware resources within the second portion.

Abstract: The semiconductor memory device includes plural core chips that are allocated with different chip identification information from each other and an interface chip that controls the plural core chips. A bit number of external unit data that is simultaneously input and output between an external device and the interface chip changes in the interface chip, and the interface chip changes chip selection information for comparison with the chip identification information, according to the bit number of the external unit data. As a result, the page configuration does not need to be changed, when the I/O configuration is changed.

Abstract: Functionality can be implemented in a virtual memory manager (VMM) to allow small pages (e.g., 4 KB) to be coalesced into large pages (e.g., 64 KB), so that a single free list can be maintained for the large pages (“maintained pages”). When a process requests a small page, the VMM can associate a maintained page with a memory segment accessible by the process. Then, the maintained page can be divided to form a set of small pages (“fragments”). The fragments can become available pages in a broken page list. The VMM can satisfy the request by allocating one of the fragments in the broken page list. If the process requests additional small pages, the additional requests can be satisfied from the broken page list. When the process terminates, the fragments in the broken page list become a maintained page and can be returned to the free list.

Abstract: The invention provides a data storage device. In one embodiment, the data storage device comprises a storage medium, a random access memory, and a controller. The storage medium stores a plurality of link tables. The random access memory comprises a plurality of storage units respectively corresponding to a plurality of logical address ranges. The controller receives a target logical address from the host, determines a target link table corresponding to a logical address set comprising the target logical address, determines a target storage unit corresponding to a logical address range comprising the target logical address, determines whether the target storage unit has stored the target link table, and when the target storage unit has stored the target link table, determines a target physical address mapped to the target logical address according to a mapping relationship stored in the target link table, and accesses data stored in the storage medium according to the target physical address.

Abstract: One embodiment of the present invention sets forth a technique for computing dynamic random access memory (DRAM) addresses from linear physical addresses for memory subsystems implementing integral power of two virtual page sizes, and an arbitrary number of available partitions. Each DRAM address comprises a row address, column address, bank address, and partition address. The linear physical address is used to generate to the DRAM address in units of a DRAM bank size. Address scrambling may be implemented to overcome transient access contention to specific DRAM pages by multiple client modules.

Abstract: A single instruction, multiple data (“SIMD”) computer system includes a central control unit coupled to 256 processing elements (“PEs”) and to 32 static random access memory (“SRAM”) devices. Each group of eight PEs can access respective groups of eight columns in a respective SRAM device. Each PE includes a local column address register that can be loaded through a data bus of the respective PE. A local column address stored in the local column address register is applied to an AND gate, which selects either the local column address or a column address applied to the AND gate by the central control unit. As a result, the central control unit can globally access the SRAM device, or a specific one of the eight columns that can be accessed by each PE can be selected locally by the PE.

Abstract: A memory device comprises a memory cell array comprising a plurality of memory blocks each comprising a plurality of memory cells and a control setting circuit. The control setting circuit divides the memory blocks into at least first and second groups based on whether each of the memory blocks comprises at least one substandard memory cell, and sets individually control parameters of the first and second groups. The substandard memory cells are identified based on test results of the memory cells with respect to at least one of the control parameters. Each memory block in the first group comprises at least one substandard memory cell, and each memory block in the second group comprises no substandard memory cell.

Abstract: Access arbiters are used to prioritize read and write access requests to individual memory banks in DRAM memory devices, particularly fast cycle DRAMs. This serves to optimize the memory bandwidth available for the read and the write operations by avoiding consecutive accesses to the same memory bank and by minimizing dead cycles. The arbiter first divides DRAM accesses into write accesses and read accesses. The access requests are divided into accesses per memory bank with a threshold limit imposed on the number of accesses to each memory bank. The write receive packets are rotated among the banks based on the write queue status. The status of the write queue for each memory bank may also be used for system flow control. The arbiter also typically includes the ability to determine access windows based on the status of the command queues, and to perform arbitration on each access window.

Abstract: A nonvolatile memory system is described with novel architecture coupling nonvolatile storage memory with random access volatile memory. New commands are included to enhance the read and write performance of the memory system.

Abstract: Various embodiments include apparatus, systems, and methods having multiple dice arranged in a stack in which a defective cell may be replaced by a spare cell on the same die or a different die. Other embodiments are described.

Abstract: In one embodiment, the present invention includes a processor that has an on-die storage such as a static random access memory to store an architectural state of one or more threads that are swapped out of architectural state storage of the processor on entry to a system management mode (SMM). In this way communication of this state information to a system management memory can be avoided, reducing latency associated with entry into SMM. Embodiments may also enable the processor to update a status of executing agents that are either in a long instruction flow or in a system management interrupt (SMI) blocked state, in order to provide an indication to agents inside the SMM. Other embodiments are described and claimed.

Abstract: A plurality of data slices are generated from a block of data to be stored in the dispersed storage system. A plurality of dispersed storage units are determined for storing the plurality of data slices, based on a geographical location associated with the plurality of dispersed storage units.

Abstract: Disclosed is an apparatus to reduce broadcasts in multiprocessors including a plurality of processors; a plurality of memory caches associated with the processors; a plurality of translation lookaside buffers (TLBs) associated with the processors; and a physical memory shared with the processors memory caches and TLBs; wherein each TLB includes a plurality of entries for translation of a page of addresses from virtual memory to physical memory, each TLB entry having page characterization information indicating whether the page is private to one processor or shared with more than one processor. Also disclosed is a computer program product and method to reduce broadcasts in multiprocessors.

Abstract: Apparatuses and methods are illustrated that relate to a web application running on a server accessible from the Internet. Aspects of the disclosure relate to an apparatus for outputting presentation code in response to receiving at least a page identifier. The apparatus comprises an electronic processor, a tangible medium storing a mapping table, template file, presentation and business rules, and computer-executable instructions.

Abstract: A memory card is attached to a host device, and includes a data control circuit which transfers data with respect to the host device in synchronism with a rise edge and a fall edge of a clock signal.

Abstract: A memory interface device has a write detection section that detects the write of a predetermined unit amount of data by a memory write unit into a memory. A signal generation section generates a signal to notify the memory write unit that readout of data from the memory by a memory readout unit has been completed. A data storage amount measurement device measures an amount of data stored in the memory during the memory write procedures. A memory readout control section generates an interrupt signal with respect to the memory readout unit when the stored data amount in the memory reaches a predetermined readout start storage amount. A timer counts a period in which writing of the predetermined unit amount of the data into the memory by the memory write unit is discontinued and outputs a timeout signal to the memory readout control section when a value of the period count reaches a predetermined timer period.

Abstract: A multipath accessible semiconductor memory device having a mailbox area and a mailbox access control method thereof are provided. The semiconductor memory device includes N number of ports, at least one shared memory area allocated in a memory cell array, and N number of mailbox areas for message communication. The at least one shared memory area is operationally connected to the N number of ports, and is accessible through a plurality of data input/output lines to form a data access path between the at least one shared memory area and one port, having an access right to the at least one memory area, among the N number of ports. The N number of mailbox areas are provided in one-to-one correspondence with the N number of ports and are accessible through the plurality of data input/output lines when an address of a predetermined area of the at least one shared memory area is applied to the semiconductor memory device. An efficient layout of mailboxes and an efficient message access path can be obtained.

Abstract: A technique for generating a list of all N-bit unsigned binary numbers by starting with an initial number less than some power of 2, successively multiplying the number by that power of 2 and adding the largest non-negative number less than that power of 2 such that the new number is not a duplicate of any of those already generated, and using the resulting lists to generate efficient hashing and serial decoding hardware and software.

Abstract: A memory card is attached to a host device, and includes a data control circuit which transfers data with respect to the host device in synchronism with a rise edge and a fall edge of a clock signal.

Abstract: A method of assigning a physical address to a tape-based data storage device is provided. The method includes receiving a first initialization signal from a system controller at an input port associated with a first tape-based data storage device and prohibiting communication at an output port associated with the first tape-based data storage device. The method further includes providing a first confirmation signal to the system controller in response to receiving the first initialization signal and receiving an instruction from the system controller to enable communication at the output port associated with the first tape-based data storage device in response to the first confirmation signal. The method also includes determining a physical address associated with said first tape-based data storage device based on the instruction from the system controller, the physical address enabling communication at the output port of the first tape-based storage device.

Abstract: An IOMMU for controlling requests by an I/O device to a system memory of a computer system includes control logic and a cache memory. The control logic may translate an address received in a request from the I/O device. If the request includes a transaction layer protocol (TLP) packet with a process address space identifier (PASID) prefix, the control logic may perform a two-level guest translation. Accordingly, the control logic may access a set of guest page tables to translate the address received in the request. A pointer in a last guest page table points to a first table in a set of nested page tables. The control logic may use the pointer in a last guest page table to access the set of nested page tables to obtain a system physical address (SPA) that corresponds to a physical page in the system memory. The cache memory stores completed translations.

Abstract: An integrated circuit memory device has a memory array and control logic with at least a first addressing mode in which the instruction includes a first instruction code and an address of a first length; and a second addressing mode in which the instruction includes the first instruction code and an address of a second length. The first length of the address is different from the second length of the address.

Abstract: A serial flash memory and an address transmission method thereof. The serial flash memory selectively addresses a first memory space according to a first address length or addresses a second memory space according to a second address length longer than the first address length. If the first memory space is addressed according to the first address length, a first memory address is completely received within an address time duration so that data corresponding to the first memory address is initially outputted from a starting clock. In the address transmission method, if the second memory space is addressed according to the second address length, a portion of a second memory address is received within the address time duration. The other portion of the second memory address is received within a waiting time duration so that data corresponding to the second memory address is initially outputted from the starting clock.

Abstract: A graphic accelerator including a frame memory and the same interface as a memory of a processor and a graphic accelerating method are provided. The graphic accelerator includes: a frame memory; an accelerator controller having the same interface as a memory of the processor on an input side and recording data, which should be transmitted from the processor to the display device, in the frame memory; and a display DMA (Direct Memory Access) transmitting the data recorded in the frame memory to the display device in a DMA manner. A memory bandwidth of the processor is not reduced even by continuous reading operations based on DMA transmission of the display device, by recording data corresponding to the frame memory in the graphic accelerator disposed outside the processor.

Abstract: In an embodiment, a memory management unit (MMU) is configured to retain a block of data that includes multiple page table entries. The MMU is configured to check the block in response to TLB misses, and to supply a translation from the block if the translation is found in the block without generating a memory read for the translation. In some embodiments, the MMU may also maintain a history of the TLB misses that have used translations from the block, and may generate a prefetch of a second block based on the history. For example, the history may be a list of the most recently used Q page table entries, and the history may show a pattern of access that are nearing an end of the block. In another embodiment, the history may comprise a count of the number of page table entries in the block that have been used.

Abstract: Methods and arrangements for remapping the map between logical space and physical space in non-volatile storage are described. Embodiments include transformations, code, state machines or other logic to divide the non-volatile storage of the computing device into two portions, a fixed portion and a floating portion. The embodiments may also include remapping in system firmware of the computing device the current map from logical space to physical space of the floating portion of the non-volatile storage. The embodiments may also include storing the revised map. The embodiments may also include using the revised map to access the floating portion of the non-volatile storage.