Abstract: Embodiments of the present disclosure relate to a memory system and an operating method of the memory system. According to embodiments of the present disclosure, the memory system may, with respect to data stored in the memory device, write the data to a first target memory block or a second target memory block among the plurality of memory blocks according to whether a data type of the data is a read-intensive type or a write-intensive type.

Abstract: Systems and methods are provided to enable parallelized multiply-accumulate operations in a systolic array. Each column of the systolic array can include multiple busses enabling independent transmission of input partial sums along the respective bus. Each processing element of a given columnar bus can receive an input partial sum from a prior element of the given columnar bus, and perform arithmetic operations on the input partial sum. Each processing element can generate an output partial sum based on the arithmetic operations, provide the output partial sum to a next processing element of the given columnar bus, without the output partial sum being processed by a processing element of the column located between the two processing elements that uses a different columnar bus. Use of columnar busses can enable parallelization to increase speed or enable increased latency at individual processing elements.

Abstract: A computer system, processor, programming instructions and/or method for managing operations of a gather buffer for a processor core load storage unit. The processor core includes a processing pipeline having one or more execution units for processing unaligned load instructions that executes in two phases to satisfy. A buffer storage element is provided having a plurality of entries for temporarily collecting partial writeback results retrieved from the memory that are associated with first phase accesses for each of a plurality of unaligned load instructions. An associated logic controller device tracks two parts of the unaligned load to be gathered at independent times, wherein said partial result stored at said buffer storage element comprises a first part of an unaligned load. The second phase load access for the same instruction is independently accessed and later merged with first part of the load data at byte granularity to satisfy the load.

Abstract: This application discloses a method and an apparatus for processing information, a communications device, and a communications system. The communications device is configured to obtain a starting position of an output bit sequence in a coded block in a circular buffer, and determine the output bit sequence in the coded block based on a length of the output bit sequence and the starting position. A value of the starting position is one of {p0, p1, p2, . . . , pkmax?1}, where 0?pk<NCB, pk is an integer, k is an integer, 0?k<kmax, NCB is a size of the coded block, and kmax is an integer greater than or equal to 4.

Abstract: With respect to a search method, for execution by a computer, for searching for multidimensional data that satisfies a search condition in a table representing a multidimensional data set including a plurality of search target parameters, the search method includes creating multidimensional indexes including a plurality of transposition blocks for identifying record numbers based on values of parameters of the plurality of search target parameters by using the multidimensional data set included in the table, and identifying a target record number of the multidimensional data that satisfies the search condition in the multidimensional data set included in the table by using the multidimensional indexes.

Abstract: In one embodiment, a system includes a non-volatile memory that may serve as both the main memory system and the backing store (or persistent storage). In some embodiments, the non-volatile memory is divided into a main memory portion and a persistent portion. Data in the main memory operation may be encrypted using one or more first keys, and data in the persistent portion may be encrypted using one or more second keys, in an embodiment. The volatile behavior of main memory may be implemented by discarding the one or more first keys in a power down event or other event that indicates a loss of main memory data, while the one or more second keys may be retained. In one embodiment, the physical address space of the non-volatile memory may be a mapping from a second physical address space that is used within the system.

Abstract: In various examples, a VPU and associated components may be optimized to improve VPU performance and throughput. For example, the VPU may include a min/max collector, automatic store predication functionality, a SIMD data path organization that allows for inter-lane sharing, a transposed load/store with stride parameter functionality, a load with permute and zero insertion functionality, hardware, logic, and memory layout functionality to allow for two point and two by two point lookups, and per memory bank load caching capabilities. In addition, decoupled accelerators may be used to offload VPU processing tasks to increase throughput and performance, and a hardware sequencer may be included in a DMA system to reduce programming complexity of the VPU and the DMA system. The DMA and VPU may execute a VPU configuration mode that allows the VPU and DMA to operate without a processing controller for performing dynamic region based data movement operations.

Abstract: A data storage system includes a storage medium including a plurality of memory cells; a storage controller in communication with the storage medium; and an electrical interface between the storage medium and the storage controller. The electrical interface includes an N-bit data bus; a data strobe; a command latch enable signal; and an address latch enable signal; wherein, while the command latch signal or the address latch enable signal is asserted, the storage medium is configured to: (i) receive command or address data via a subset of lines of the data bus; and (ii) latch the command or address data using the data strobe.

Abstract: Data requests can be serviced by multiple paths of smart network interface cards (NICs). For example, a system can receive a request for data at a first path of a smart NIC. The first path can be a hardware-implemented path. The system can send one or more parameters of the request to a second path of the smart NIC. The second path can be a slower path than the first path and configured to execute a routing algorithm for the request. The system can receive routing information for the request from the second path based on the routing algorithm and transmit the request to a storage node based on the routing information.

Abstract: A serial peripheral interface (SPI) communication system includes a memory configured with a start register address and an end register address that define a register address range for a data operation; a chip select terminal configured to receive a chip select signal comprising an active and idle signal levels that define a plurality of chip select frames; a serial data input terminal configured to receive a master out, slave in (MOSI) signal, wherein the MOSI signal includes configuration information received in a first chip select frame of the data operation, wherein the configuration information includes an operation command bit indicating whether the data operation is a write operation or a read out operation and an auto-incrementation control bit indicating whether automatic register address incrementation across chip select frames is enabled or disabled; and a serial data output terminal configured to transmit a master in, slave out (MISO) signal.

Abstract: Techniques for generating an enhanced backup catalog that preserves file extents for a file that is being backed up are disclosed. A file is identified. A directory structure is consulted to identify a mapping of extents that exist for the file. A backup process is triggered to backup the file. A backup catalog, which is generated as a part of the backup process, is modified to include a list of the extents for the file. This modified backlog catalog is referred to as an enhanced backup catalog. The enhanced backup catalog is then stored in remote backup storage.

Abstract: Described herein is a system and method of performing flow control for multi-threaded access to contentious resource(s) (e.g., shared memory). A request to enter a critical section of code by a particular thread of a plurality of concurrent threads is received. A determination is made as to whether or not to allow the particular thread to enter the critical section of code based, at least in part, upon a CPU core associated with the particular thread, a state associated with the particular thread, and/or a processing rate in the critical session of code associated with the particular thread. When it is determined to allow the particular thread to enter the critical section of code, the particular thread is allowed to enter the critical section of code.

Abstract: A semiconductor memory device includes a plurality of planes defined in a plurality of chip regions; and a rescue circuit configured to disable a failed plane and enable a normal plane from among the plurality of planes, wherein the semiconductor memory device operates with only normal planes that are enabled.

Abstract: A register mapping circuit for recovering a register mapping state associated with a flushed instruction by traversing ROB entries from a snapshot of another register mapping state. The register mapping circuit includes a ROB control circuit, a snapshot circuit, and a register rename recovery circuit (RRRC). The ROB control circuit allocates ROB entries to instructions entering a processor pipeline, including a target ROB entry allocated to a target instruction and other ROB entries allocated to other instructions. The snapshot circuit captures snapshots of logical register-to-physical register mapping states in the rename map table in association with a subset of instructions that could be flushed. If the target instruction is flushed, the RRRC restores the rename map table register mapping state corresponding to the target instruction based on a snapshot in a ROB entry allocated to another instruction, and traverses register mapping updates in the intervening ROB entries.

Abstract: In an example, a system includes network nodes implementing an object store. The system may determine reference count updates for objects stored on the object store, and delay persisting the reference count updates to a persistent storage. The system may cancel reference count updates that increments and decrement between the determined reference count update and another reference count update received during the delay in persisting the reference count update to the persistent storage to minimize writing to the object record.

Abstract: A storage system comprises a plurality of storage devices and an associated storage controller. The storage controller is configured to receive a request to copy a first range of logical addresses to a second range of logical addresses, determine at least one physical block of the storage devices to which the first range of logical addresses is mapped, map the second range of logical addresses to the determined at least one physical block, and add at least one content-based signature associated with the determined at least one physical block to a pending increment data structure that includes content-based signatures corresponding to physical blocks for which an increment of an associated reference count is pending. The storage controller is further configured to execute a pending increment of a reference count associated with a given physical block corresponding to at least one of the content-based signatures in the pending increment data structure.

Abstract: A system includes a memory with a ring buffer having a plurality of slots and a producer and consumer processor in communication with the memory. The producer processor is configured to receive a new memory entry and detect a failure to produce the new memory entry to a slot in the ring buffer. Each memory entry in the ring buffer has an entry structure to maintain a list of extra entries. The producer processor is also configured to determine a location of an entry pointer for a last produced memory entry in the ring and add the new entry to the list of extra entries in the respective slot in the ring. Responsive to consuming the last produced memory entry, the consumer processor is configured to check whether the last produced memory entry includes any other memory entries in the list of extra entries and consume the new memory entry.

Abstract: In an example, a system includes network nodes implementing an object store. The system may determine reference count updates for objects stored on the object store, and delay persisting the reference count updates to a persistent storage. The system may cancel reference count updates that increments and decrement between the determined reference count update and another reference count update received during the delay in persisting the reference count update to the persistent storage to minimize writing to the object record.

Abstract: One embodiment relates to a memory structure that includes a bank group and a port emulation circuit module. The bank group includes a plurality of memory banks, each memory bank having one read port and one write port. The port emulation circuit module provides a group read/write port and a group read port for the bank group. Another embodiment relates to a port emulation circuit module. The port emulation circuit module includes a port emulation control circuit that receives control signals including a first address for a group read/write port and a second address for a group read port, a first data path circuit for the group read/write port, and a second data path circuit for the group read port, wherein the second data path circuit outputs a second read data. Other embodiments and features are also disclosed.

Abstract: A network media delivery system includes client devices and a host device. Each client device has a network interface, an engine for processing media data, and a media interface. The host device, which can be a computer, establishes network communication links with the client devices, which can be networked media stations, and sends media data to the client devices. The media data can be sent wirelessly as packets of media data transmitted at intervals to each client device. In one embodiment, the host device controls processing of media data such that processed media is delivered in a synchronized manner at each of the client devices. In another embodiment, the host device controls processing of media data such that processed media is delivered in a synchronized manner at the host device and at least one client device.

Abstract: Current tasks being executed in a set of modules of a signal processing system managed via an interface block are aborted so as to permit the execution of new tasks by pipelining eliminating transactions of said current tasks and executing transactions of the new tasks. Upon arrival of a signal to abort the current tasks, data and/or memory accesses present in said interface block are discarded.

Abstract: A memory server providing remote memory for servers independent from the memory server. The memory server includes memory modules and a page table. A memory controller for the memory server allocates memory in the memory modules for each of the servers and manages remote memory accesses for the servers. The page table includes entries identifying the memory module and locations in the memory module storing data for the servers.

Abstract: The present disclosure provides a processor, and associated method, for performing parallel processing within a register. An exemplary processor may include a processing element having a compute unit and a register file. The register file includes a register that is divisible into lanes for parallel processing. The processor may further include a mask register and a predicate register. The mask register and the predicate register respective include a number of mask bits and predicate bits equal to a maximum number of divisible lanes of the register. A state of the mask bits and predicate bits is set to respectively achieve enabling/disabling of the lanes from executing an instruction and conditional performance of an operation defined by the instruction. Further, the processor is operable to perform a reduction operation across the lanes of the processing element and/or generate an address for each of the lanes of the processing element.

Abstract: A memory storage system that includes at least a storage controller, a first non-volatile, solid-state memory and a second non-volatile, solid-state memory. The storage controller has an interface to receive commands from a host system. The first non-volatile, solid-state memory device is coupled with the storage controller to at least store data received from the host system. The second non-volatile, solid-state memory is coupled with the storage controller to store context information corresponding to the data stored in the first non-volatile, solid-state memory device.

Abstract: In an embodiment of the invention, a method of performing a byte-rotate-merge on computer hardware is discussed. Byte-rotating is performed on first and second source operands and are byte-rotated by first and second rotation constants respectively. The first byte-rotate output and the second byte-rotate output are merged. Bytes from the first byte-rotate output are output to a byte-rotate-merge output when control bits from a byte-mask are logical ones. Bytes from the second byte-rotate output are output to a byte-rotate-merge output when control bits from the byte-mask are logical zeros.

Abstract: The disclosure is related to optimizing generation of intermediate representation (IR) for a script code by eliminating redundant reference count code from the IR. The reference count code includes code that manages a reference count of an object, e.g., code that increments a reference count of the object (“incref code”) and an observer code which consumes or the execution of which depends on the reference count of the object. The IR is analyzed to identify redundant reference count code. Counters associated with the object are evaluated and upon satisfying the optimization criterion, the incref code is moved closer to the observer code. The incref code and the observer code that are adjacent to each other are identified as redundant code pair and the code pair is eliminated from the IR to generate an optimized IR. The optimized IR is further converted to an executable code.

Abstract: A digital rack interface pod (DRIP) that is able to establish a communications link between a remote access appliance and a server having a USB port and a video port, to facilitate a keyboard/video/mouse (KVM) session between the server and the appliance. The DRIP is also able to establish a communications link between the appliance and an Ethernet port of the server that is associated with a service processor (SP) of the server. In this manner the DRIP is able to route data from the SP of the server to the appliance as well as communications from the appliance to the SP Ethernet port of the server. The DRIP is able to intelligently determine which packets of information received from the appliance are intended for the SP Ethernet port of the server and routes those packets to the SP Ethernet port of the server.

Abstract: In general, this disclosure describes techniques for selecting a memory channel in a multi-channel memory system for storing data, so that usage of the memory channels is well-balanced. A request to write data to a logical memory address of a memory system may be received. The logical memory address may include a logical page number and a page offset, where the logical page number maps to a physical page number and the logical memory address maps to a physical memory address. A memory unit out of a plurality of memory units in the memory system may be determined by performing a logical operation on one or more bits of the page offset and one or more bits of the physical page number. The data may be written to a physical memory address in the determined memory unit in the memory system.

Abstract: A functional unit is provided which allows for fast, parallel data read, write, and manipulation operations. The functional unit includes first and second source registers for receiving first and second data items to be processed by the functional unit, a plurality of memory tables, a combinational logic circuit, and a decoder. Each of the tables is indexed by an index comprising a portion of the first data item received by the first source register. The combinational logic circuit receives lookup results, and processes the lookup results and the second data item in the second source register to produce a result data item. The decoder circuit extracts an operational code from an instruction supplied to the functional unit, decodes the operational code, and controls the combinational logic circuit in accordance with the operational code.

Abstract: Memory access in data processing is provided using a time-based technique in which memory locations are mapped to respectively corresponding periods of time during which they are made available for access.

Abstract: An apparatus having a memory and a controller is disclosed. The controller may be configured to (i) receive a read request from a processor, the read request comprising a first value and a second value, (ii) where the read request is an indirect memory access, (a) generate a first address in response to the first value, (b) read data stored in the memory at the first address and (c) generate a second address in response to the second value and the data, (iii) where the read request is a direct memory access, generate the second address in response to the second value and (iv) read a requested data stored in the memory at the second address.

Abstract: An apparatus and associated methodology providing a data storage system including a memory having a first addressable storage space and a second differently addressable storage space. A controller selectively accesses stored instructions that when executed store a first amount of a user data set to the first addressable storage space and store a different second amount of the user data set to the second addressable storage space. The controller subsequently calculates an address increment between the stored first and second amounts, and then shifts one of the stored first and second amounts by the address increment.

Abstract: A system for writing data includes a memory, at least one memory controller and control logic. The memory stores data units. The memory controller receives a write request associated with a data unit and stores the data unit in the memory. The memory controller also transmits a reply that includes an address where the data unit is stored. The control logic receives the reply and determines whether the address in the reply differs from an address included in replies associated with other memory controllers by a threshold amount. When this occurs, the control logic performs a corrective action to bring an address associated with the memory controller back within a defined range.

Abstract: Data signal items output by a radix 4n2m fast Fourier transform (“FFT”) operation may not be in the order desired for further use of those data items (e.g., they may be output in a non-natural order rather than in a desired natural order). Memory circuitry (e.g., dual-port memory circuitry) may be used in conjunction with circuitry for addressing the memory circuitry with address signals that are reordered in a particular way for each successive set of N data items. This allows use of memory circuitry with fewer data item storage locations than would otherwise be required to reorder the data items from non-natural to natural order. In particular, the memory circuitry only needs to be able to store N data items at any one time, which is more efficient memory utilization than would otherwise be possible.

Abstract: A multimode accessible storage facility (10) is described that allows block access in a block access mode and row access in a row access mode. The facility comprises—a memory unit (20) comprising a plurality of memory banks (20.0, . . . , 20.F) each having a respective bank index (0, . . . , F), —an address generator (30) for generating for each of said memory banks a rotated bank address as a function of an input address and a shift parameter, —an input vector data rotator (40) for rotating an input vector and for providing vector elements of the rotated input vector to a respective bank of the memory unit, and—an output vector rotator (50) for inverse rotating a vector comprising vector elements retrieved from respective banks of the memory unit and for providing the rotated output vector.

Abstract: A transactional memory (TM) receives a lookup command across a bus from a processor. The command includes a memory address, a starting bit position, and a mask size. In response to the command, the TM pulls an input value (IV). The memory address is used to read a word containing multiple result values (RVs) and multiple threshold values (TVs) from memory. A selecting circuit within the TM uses the starting bit position and mask size to select a portion of the IV. The portion of the IV is a lookup key value (LKV). The multiple TVs define multiple lookup key ranges. The TM determines which lookup key range includes the LKV. A RV is selected based upon the lookup key range determined to include the LKV. The lookup key range is determined by a lookup key range identifier circuit. The selected RV is selected by a result value selection circuit.

Abstract: An arithmetic logic and shifting device is disclosed and includes an arithmetic logic unit that has a first input to receive a first operand from a first register port, a second input to receive a second operand from a second register port, and an output to selectively provide a memory address to a memory unit in a first mode of operation and to selectively provide an arithmetic output in a second mode of operation. Further, the arithmetic logic and shifting device includes a programmable shifter device that has a first input to receive data from the memory unit, a second input to receive the arithmetic output, a third input to receive an operation code of a computer execution instruction, and a shifted output to provide shifted data.

Abstract: In response to a memory allocation request received from an application thread, a random number is obtained (e.g., from a random number list previously populated with multiple random numbers). A starting location in at least a portion of a bitmap associated with a region including multiple blocks of the memory is determined based on the random number. A portion of the bitmap is scanned, beginning at the starting location, to identify a location in the bitmap corresponding to an available block of the multiple blocks, and an indication of this available block is returned to the application thread.

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: A memory device can include a data path that includes a first-in-first-out circuit (FIFO) to transfer data according to a latency between at least one memory cell array and signal connections of the memory device, the latency corresponding to a number of cycles of a periodic clock; and a self-timed section configured to transfer data independent of the clock. In addition or alternatively, a memory device can include at least one memory cell array; and a FIFO configured to transfer data between at least one memory cell array and other portions of the memory device according to a periodic clock signal, FIFO introducing a latency into the data according to a control signal generated in response to an access command. Methods corresponding to the above devices and operations are also disclosed.

Abstract: Provided are a ring buffer circuit in which a data full state and a data empty state may be correctly detected without depending on whether read and write operations are synchronous or asynchronous with each other, and a control circuit for the ring buffer circuit. The ring buffer circuit includes: a read and write memory having addresses specified by N bits; a write address counter pointer and a read address counter pointer which are provided for the read and write memory to count (N+1)-bit gray codes; and write and read address converter circuits provided to convert the (N+1)-bit gray codes output from the write and read address counter pointers into N-bit addresses which may be directly designated as write and read addresses of the read and write memory.

Abstract: The present invention provides techniques for managing ownership (i.e., control) of one or more memory buffer (mbuf) data structures within a network subsystem and a storage subsystem of a storage operating system implemented in a storage system. When the storage system receives data to be written to a storage medium, the network subsystem stores the received data in one or more variable-length chains of mbufs. Unlike conventional approaches, the received data is not subsequently copied out of the mbufs into fixed-sized data buffers for use by the storage subsystem. Instead, the storage subsystem can directly manipulate the received data stored in the mbufs. By eliminating the steps of copying data out of the mbufs and into fixed-sized data buffers, the invention reduces the amount of time and system resources consumed by the storage system when writing blocks of received data to disk storage.

Abstract: According to one embodiment, an information processing apparatus includes an information processing apparatus main body and a storage drive which is accommodated in the information processing apparatus body. The information processing apparatus main body includes a main control module which controls power activation of the storage drive based on a power activation operation by a power operation section. The storage drive includes a storage memory including a plurality of storage areas to and from which information can be written and read, a counter in which a count is incremented upon power activation, and a memory control module which stores a content of an access request in a storage area of the storage memory determined based on the count of the counter when the access request to the storage memory is made.

Abstract: A functional unit for a microprocessor is provided, which allows for fast, parallel data read, write, and manipulation operations in the microprocessor that are useful for a number of software applications, such as cryptography. The functional unit includes first and second source registers for receiving first and second data items to be processed by the functional unit, first and second banks of memory tables, a combinational logic circuit, and a decoder. The first and second banks of memory tables are in communication with the first source register, and each of the tables is indexed by an index comprising a portion of the first data item received by the first source register. Each index points to a lookup result in a respective one of the memory tables.

Abstract: A system and a method for configuring a memory controller that communicates with a memory device muxes selected pins for the data transfer. The memory controller includes a set of pins where each pin of the set is associated with a data bit and an address bit. A programmable logic block is connected to the set of pins and uses a subset of the set of pins to enable data transfer between the memory device and the memory controller depending on the size of the memory device such that the pins not included in the subset are available for other applications.

Abstract: A nonvolatile memory device includes first and second registers configured to store parameters received via an input/output (IO) unit, a microcontroller configured to control an operation of the nonvolatile memory device according to the parameter stored in the first register, and a control logic unit configured to, when a parameter is received via the IO unit while the microcontroller performs an internal operation, store the received parameter in the second register.

Abstract: An interface includes a controller that divides a burst access command into a plurality of command cycles and supplies the plurality of command cycles to a storage device including a plurality of blocks, and a block address converter that outputs an address at a first command cycle of the plurality of command cycles. The address is obtained by shifting at least one bit of an external block address input in response to the burst access command. The address is supplied to the storage device at the first command cycle, and the external block address is supplied to the storage device at a command cycle other than the first command cycle.

Abstract: A burst address generator includes a burst bit counter for receiving at least one burst bit, and increasing or decreasing the at least one burst bit, a burst bit splitter for receiving the increased or decreased at least one burst bit from the burst bit counter, and dividing the increased or decreased at least one burst bit into an X burst bit and a Y burst bit, and a selector for receiving an X address, a Y address, the X burst bit, and the Y burst bit, and generating an X burst address based on the X address and the X burst bit and a Y burst address based on the Y address and the Y burst bit.

Abstract: An address generation system and method is provided for internally storing and thereafter producing an address to be sent to a memory device. The address that is stored need not be sent from an external address bus at each clock cycle, but the processing can remain internal to the memory device. The burst-block starting address can be stored in the mirror register and output from a selector circuit, such as a multiplexer, when that address is chosen. Otherwise, the multiplexer can simply perform its normal operation of selecting between an address pointed to by a counter, the external address, or the incremented counter output, based on the state of the external counter control signals. The system includes a mirror register, a counter, and a multiplexer that selects either the mirror register stored address or the internally processed address.

Abstract: A reproducing apparatus and method includes a reproducing unit to reproduce mainstream data and sub audio data separately added in the mainstream data, wherein the reproducing unit comprises a counter used in reproducing the sub audio data. Accordingly, it is possible to more naturally reproduce still image data, such as a browsable slide show, to which sub audio data is additionally included, thus preventing an interruption in reproduction of the sub audio data even during a forward or reverse play.