Abstract: A method and apparatus for synchronizing input/output commands is provided. An incoming command mask representing an incoming input/output command associated with a memory region is created. In response to a determination that a pending input/output command associated with the memory region is pending, a bitwise inversion operation is performed on the incoming command mask to form a modified incoming command mask. A bitwise AND operation is performed on the modified incoming command mask and the pending command mask to form a pending command locking mask associated with the pending input/output command. A bitwise OR operation is performed between an existing memory lock for a same type of commands and incoming command bit mask to form a new memory region lock.

Abstract: A memory controller converts controller output signals output from a controller into memory input signals according to the operation specifications of memory chips to operate, and outputs the resultant to the memory chips through a common bus. The memory controller also receives memory output signals output from the memory chips through the common bus, and converts the received signals into controller input signals receivable to the controller. This allows the single memory controller to access the plurality of types of memory chips. As a result, the memory controller can be reduced in chip size, lowering the cost of the memory system.

Abstract: The present invention extends to methods, systems, and computer program products for maintaining a count for lock-free stack access. A numeric value representative of the total count of nodes in a linked list is maintained at the head node for the linked list. Commands for pushing and popping nodes appropriately update the total count at a new head node when nodes are added to and removed from the linked list. Thus, determining the count of nodes in a linked list is an order 1 (or O(1)) operation, and remains constant even when the size of a linked list changes.

Abstract: Opportunistically locked files are accessed in real-time from a kernel mode file system filter driver without breaking opportunistic locks. A file system is monitored, and the granting of an opportunistic lock to a specific file object referring to a specific underlying file is detected. The granting of the detected opportunistic lock results in both the file object and the underlying file being opportunistically locked. A reference to the opportunistically locked file object is cached. The file system filter driver filters a request to access the opportunistically locked file via a second file object which is not opportunistically locked. The kernel mode file system filter driver uses the cached reference to access the opportunistically locked underlying file. This enables access of the opportunistically locked file from the kernel mode file system filter driver without breaking the opportunistic lock.

Abstract: A storage device, enabling elimination of redundant write operations of non-selected data and enabling optimization of arrangement of pages to a state efficient for rewriting, having two flash memories which can be accessed in parallel, a page register for acquiring data in parallel from the flash memories and temporarily storing the same, and a control circuit having a built-in RAM in which is constructed an address conversion table for managing correspondence between logical addresses and physical addresses in units of data stored in parallel in the page register, wherein data is rewritten by updating of the address conversion table and additional writing into a storage medium.

Abstract: Provided is a memory system that has a plurality of memory banks and a plurality of link controllers. For each memory bank, there is first switching logic for receiving output for each link controller, and for passing on the output of only one of the link controllers to the memory bank. For each link controller, there is second switching logic for receiving an output of each memory bank, and for passing on the output of only one of the memory banks to the link controller. According to an embodiment of the invention, there is switch controller logic for controlling operation of both the first switching logic and the second switching logic to prevent simultaneous or overlapping access by multiple link controllers to the same memory bank, and for preventing simultaneous or overlapping access to multiple banks by the same link controller.

Abstract: Methods and apparatus are disclosed for reducing write-to-read turnaround times using shadow writes in memory controllers and in DRAM. Embodiments of controllers including shadow write control logic may, in response to receiving a write request, issue an external write column address strobe (CAS) to DRAM to latch a valid write CAS address, and assert a set of write data values to be stored in a set of DRAM locations corresponding to the write CAS address. After asserting the write CAS and prior to asserting the complete set of write data values, such memory controllers may, in response to receiving a read request, issue an external read CAS to DRAM to indicate a valid read CAS address. A set of read data values from a second set of DRAM locations corresponding to the read CAS address, are received with reduced turnaround time after asserting the complete set of write data values.

Abstract: A memory-use-information memory area stores therein a program ID, a request-source memory address, a request memory size which configure information for uniquely identifying a program file loaded into a storage area for virtual machine-A or storage area for virtual machine-B in association with a physical memory address. A memory reservation section uses, as the retrieval key, the program ID, request-source memory address, and request memory size of a program file corresponding to a memory reservation request to retrieval the memory-use-information memory area. When a entry that matches said retrieval key exists, the memory reservation section allows sharing of the memory area between a plurality of virtual machines.

Abstract: A solid-state drive, a solid-state drive access unit allocation/data storage approach, and a solid-state drive access unit access/data retrieval approach are described that improve the efficiency with which data, that has been stored to the solid-state drive in association with a series of logical block addresses, can be retrieved from the solid-state drive. The described access unit allocation approach assures that data stored in the solid-state drive in association with a sequential series of logical block addresses is stored and maintained in solid-state drive access units, i.e., addressable units of solid-state drive memory, that allow parallel read access to the data via parallel memory access I/O channels internal to the solid-state drive. In this manner, the time required to retrieve data associated with a sequential series of logical block addresses from corresponding access units within the solid-state drive is reduced.

Abstract: A hierarchical memory request stream arbitration technique merges coherent memory request streams from multiple memory request sources and arbitrates the merged coherent memory request stream with requests from a non-coherent memory request stream. In at least one embodiment of the invention, a method of generating a merged memory request stream from a plurality of memory request streams includes merging coherent memory requests into a first serial memory request stream. The method includes selecting, by a memory controller circuit, a memory request for placement in the merged memory request stream from at least the first serial memory request stream and a merged non-coherent request stream. The merged non-coherent memory request stream is at least partially based on an indicator of a previous memory request selected for placement in the merged memory request stream.

Abstract: From among a plurality of threads accessing a shared data object, one thread acquires a “master” status to arbitrate among the requests of competing threads during a given session of data access to the shared data object. During the session, the master thread resolves any conflicts resulting from attempts to access or modify the shared data object by other threads, and only the master thread may apply modifications to the shared data object during the session. Meanwhile, during the session, non-master threads may perform non-blocking read operations on the shared data object. During a subsequent session, a different thread may acquire master status.

Abstract: A method includes operating an arbitration logic of a memory controller at a core clock frequency lower than that of a memory clock frequency. The memory controller is configured to generate a command sequence including a number of commands in accordance with a number of external requests to access the memory. The method also includes parallelizing the number of commands in the command sequence based on a timing requirement for a non-first command in the command sequence defined by a memory-access protocol being satisfied at a rising edge or a falling edge of the core clock relative to a previous command in the command sequence. Further, the method includes ensuring, through the parallelizing, availability of the number of commands in the command sequence to a memory interface operating at the memory clock frequency at a command rate equal to the memory clock frequency.

Abstract: According to one embodiment, a command generator sequentially and speculatively issues channel-by-channel access commands to a memory interface in a predetermined access process. A purger returns a series of unexecuted already-issued access commands using a purge response if an error occurs in any of memory accesses via a plurality of channels. A command progress manager updates command progress information such that the command progress on each of the plurality of channels returns to a position specified in an oldest access command of a series of the returned access commands issued to the channel. The command generator issues the channel-by-channel access commands including the oldest access command to the memory interface based on the updated command progress information.

Abstract: When data is encrypted and stored for a long time, encryption key(s) and/or algorithm(s) should be updated so as not to be compromised due to malicious attack. To that end, stored encrypted data is converted in the storage system with new set of cryptographic criteria. During this process, read and write requests can be serviced.

Abstract: A memory controller includes ports and corresponding tables. Each port is receptive to one or more memory modules. Each table includes entries mapping memory addresses to the memory modules. Each entry corresponds to no more than one of the memory modules. The tables support asymmetric population of the memory modules within the ports; each port is capable of having a different number of memory modules relative to the other ports. The tables impose no restrictions on where the memory modules are to be inserted within the ports, both number-wise and position-wise. The tables are independently configurable; the configuration of each table is modifiable independently of the configurations of the other tables. Each table is dynamically configurable. The entries of a table are modifiable to reflect changes in the number and type of the memory modules connected, without restarting or temporarily halting the computer system containing the memory controller.

Abstract: A method of transmitting compressed data from a main memory to an input/output adaptor (IOA)/input/output processor (IOP), includes sending compressed memory directory information to the IOA/IOP and copying a content of the memory to the IOA/IOP using a direct memory access (DMA) operation, without decompressing the data.

Abstract: A cache memory system comprises a cache memory and a cache controller that receives a first address to access the cache memory. The cache controller includes a first address transformer receives the first address and to transform it into one first cache address; the first cache address is used for accessing the cache memory to retrieve a first part of a tag address portion. The cache controller includes a hit detector establishes an partial hit condition based on a comparison of the retrieved a first part of the tag address portion and a first predetermined part of the first address, and a second address transformer receives the first address and to transform it into one second cache address. The cache controller uses the one second cache address for accessing the cache memory to retrieve a data word in case the partial hit condition is established.

Abstract: A method for providing quality of service to a plurality of hosts accessing a common resource is described. The common resource may be a middle-tier or back-end server. A client IO request is received at one host of the plurality of hosts from one of a plurality clients executing as software entities on respective hosts. The host determines whether an issue queue is full. The IO request is issued to the common resource when the issue queue is not full. A current average latency observed at the host and an adjusted window size is calculated, based at least in part on the current average latency. The issue queue is resized to correspond with the adjusted window size.

Abstract: A memory controller mechanism is operable in a first mode and a second mode. In the first mode, a first memory controller portion of the mechanism can use a first set of data terminals to perform a first external bus access operation (EBAO) and a second memory controller portion of the mechanism can use a second set of data terminals to perform a second EBAO. The first and second EBAO operations may be narrow accesses that occur simultaneously. In the second mode, one of the controllers can use both the first and second sets of data terminals to perform a wider third EBAO. The memory controller mechanism can dynamically switch between first mode and second mode operations. In situations in which one of the sets of data terminals would not otherwise be used, performing wide accesses in the second mode using the one set of data terminals improves bus utilization.

Abstract: Systems and/or methods that facilitate transferring data between a processor component and memory components are presented. A transfer controller component facilitates controlling data transfers in part by receiving respective subsets of data from respective memory components and arranging the respective subsets of data based in part on a desired predefined data order. The processor component generates a transfer map that includes information to facilitate arranging data in a predefined order. The processor component generates respective subsets of commands that are provided to queue components in respective memory components to retrieve desired data from the respective memory components. Each memory component services the commands in its queue component in an independent and parallel manner, and transfers the data retrieved from memory to the transfer controller component, which can arrange the received data in a predefined order for transfer to the processor component.

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.

Abstract: A memory expansion blade for a multi-protocol architecture, includes dual inline memory modules (DIMMs) and a multi-protocol memory controller coupled to the DIMMs and operable to control operations of the DIMMs. The multi-protocol memory controller includes one or more memory channel controllers, with each of the memory channel controllers coupled to a single channel of DIMM, and where the DIMM in each single channel operate according to a specific protocol. The controller further includes a protocol engine coupled to the memory channel controllers, where the protocol engine is configurable to accommodate one or more of the specific protocols, and a system interface coupled to the protocol engine and configurable to provide electrical power and signaling appropriate for the specific protocols.

Abstract: In a computer system supporting memory compression, wherein memory compressed data is managed in units of memory sectors of size S, wherein data is stored on disk in a different compressed format, and wherein data on said disk is managed in units of disk sectors of size D, a method for storing memory compressed data on a compressed disk includes combining at least one of compressed memory directory information, a system header, compressed data controls, and pads into a data structure having a same size S as a memory sector, grouping the data structure and the data contained in the desired memory sectors into groups of D/S items, and storing each of the groups in a separate disk sector.

Abstract: In an access control method for a memory including a first memory block and a second memory block, first flag is read from the first memory block, and second flag is read from the second memory block. A comparison between the first flag and the second flag is done, and then, the comparison result is outputted as a first calculation value. One of the first memory block and the second memory block is selected in response to the first calculation value to decide a valid memory block. In the access control method for the memory executed in the above-mentioned manner, even when the first flag and the second flag employed in order to determine the valid memory block are brought into any statuses, the valid memory block does not become indefinite, and a single memory block can be determined as the valid memory block.

Abstract: An access method for use in a memory device is provided. The memory device comprises a data area having a plurality of data blocks and a spare area having a plurality of spare blocks. First, data from a host is received. A spare block is popped from the spare area and the received data is programmed into the popped spare block accordingly. A data block corresponding to the data is pushed to the spare area. The pushed data block is erased when the memory device is waiting for a specific instruction to be issued from the host.

Abstract: A secure virtual tape management system with early read support options. The system includes at least two mainframe hosts having a catalog storing tape related information. A primary virtual tape emulation system includes an adaptor and includes software for facilitating remote configuration and utilization of the virtual tape management. A virtual tape system catalog storing tape related information is attached to the virtual tape management. Remote data storage devices may be in communication with the virtual tape management central processing unit. Software resident on the catalog monitors tape related information on the primary virtual tape emulation system for criteria matching a virtual tape to be made available to a secondary host and initiates immediate transfer of that data allowing it to be read in a paced manner by the secondary host before the primary host has completed its series of tape writes.

Abstract: To reduce overhead of data transfer between processor cores and improve a processing capability of a processor, there is provided a processor including: a CPU for performing computing processing; an internal memory for storing data; and a data transfer unit for performing data transfer between the internal memory and a shared memory, in which: the data transfer unit includes: a command chain module for executing a command sequence formed of a plurality of commands including a data transfer instruction; and a monitor module for reading data set in advance in the internal memory and repeatedly monitoring the data until a comparative value and a value of the data become equal to each other, when one of the plurality of commands of the command sequence thus read is a predetermined command; and the command chain module executes a next command in the command sequence after the monitor module has finished monitoring.

Abstract: A computer system includes a controller coupled to a plurality of memory modules each of which includes a memory hub and a plurality of memory devices divided into a plurality of ranks. The memory hub is operable to configure the memory module to simultaneously address any number of ranks to operate in a high bandwidth mode, a high memory depth mode, or any combination of such modes.

Abstract: Methods and apparatus for programmable logic devices including embedded processors having a dual-port SRAMs. A programmable logic integrated circuit includes a programmable logic portion having a plurality of logic elements, programmably configurable to implement user-defined combinatorial or registered logic functions, and an embedded processor portion coupled to the programmable logic portion. The embedded processor portion includes a processor, and a memory block coupled to the processor. The memory block includes a first plurality of memory cells for storing data, a second plurality of memory cells for storing data, a first port coupled to the first and second pluralities of memory cells, a second port coupled to the first and second pluralities of memory cells, and an arbiter coupled to the first port and the second port.

Abstract: Subject matter disclosed herein relates to read and write processes of a memory device.

Abstract: A main LSI includes a plurality of master circuits transmitting access requests to a SDRAM, and includes an input interface receiving an access request from a master circuit in a sub LSI. Further, the main LSI includes an arbitration circuit receiving the access requests from the internal master circuits and from the input interface, sequentially selecting, in accordance with a predetermined arbitration rule, a master circuit to be allowed to access the SDRAM, and determining output timings for addresses pertaining to the data transfers from the sequentially selected master circuits. The main LSI also includes an access signal generation circuit causing the sequentially selected master circuits to access the SDRAM in accordance with the corresponding output timings.

Abstract: A memory is used by concurrent threads in a multithreaded processor. Any addressable storage location is accessible by any of the concurrent threads, but only one location at a time is accessible. The memory is coupled to parallel processing engines that generate a group of parallel memory access requests, each specifying a target address that might be the same or different for different requests. Serialization logic selects one of the target addresses and determines which of the requests specify the selected target address. All such requests are allowed to proceed in parallel, while other requests are deferred. Deferred requests may be regenerated and processed through the serialization logic so that a group of requests can be satisfied by accessing each different target address in the group exactly once.

Abstract: A portable data storage device compatible with both standard and high definition digital video cameras is provided. The device includes at least one SDI I/O, and preferably at least one audio I/O and preferably at least one medium speed I/O interface. A device controller takes the high speed serial data, packetizes it, and then sends it out to a plurality of memory modules. Preferably each memory module includes four NAND clusters, each NAND cluster consisting of a flash memory controller and two NAND flash memories. Interposed between the device controller and the memory modules are a plurality of memory controllers, each memory controller controlling a group of memory modules. A user interface is coupled to the device controller, the interface including a display capable of at least two user-selectable orientations, record/playback controls and a four-way directional control pad.

Abstract: A non-volatile memory system and a method for reading data therefrom are provided. The data comprises a first sub-data and a second sub-data. The non-volatile memory system comprises a first storage unit and a second storage unit, adapted for storing the two sub-data respectively. The first storage unit reads a first command from the controller, and stores the first sub-data temporarily as the first temporary sub-data according to the first command. The second storage unit reads a second command from the controller, and stores the second sub-data temporarily as the second temporary sub-data according to the second command. The first temporary sub-data is read from the first storage unit. Then, the first storage unit reads a third command from the controller. The second temporary sub-data is also read from the second storage unit while reading the third command. The time for reading data from the non-volatile memory system is reduced.

Abstract: In a computer system supporting memory compression and wherein data is stored on a disk in a different compressed format, and wherein an IOA (input/output adaptor)/IOP (input/output processor) selectively reads from and writes to a main memory through a direct memory access (DMA) operation, a method for transmitting compressed data from the IOA/IOP to the main memory includes reserving a set of free memory sectors to contain the data in said main memory, sending to the IOA/IOP addresses of said memory sectors, copying the data from the IOA/IOP to said memory sectors using said DMA operation, constructing at the IOA/IOP compressed memory directory information defining how and where the data is stored in memory, sending the memory directory information to a memory controller, and storing the memory directory information in the compressed memory directory structure.

Abstract: A memory system includes a master device, such as a graphics controller or processor, and an integrated circuit memory device operable in a dual column addressing mode. The integrated circuit memory device includes an interface and column decoder to access a row of storage cells or a page in a memory bank. During a first mode of operation, a first row of storage cells in a first memory bank is accessible in response to a first column address. During a second mode of operation, a first plurality of storage cells in the first row of storage cells is accessible in response to a second column address during a column cycle time interval. A second plurality of storage cells in the first row of storage cells is accessible in response to a third column address during the column cycle time interval. The first and second pluralities of storage cells are concurrently accessible from the interface.

Abstract: Systems and methods for reducing delays between successive write and read accesses in multi-bank memory devices are provided. Computer circuits modify the relative timing between addresses and data of write accesses, reducing delays between successive write and read accesses. Memory devices that interface with these computer circuits use posted write accesses to effectively return the modified relative timing to its original timing before processing the write access.

Abstract: Multiple pipelined Translation Look-aside Buffer (TLB) units are configured to compare a translation address with associated TLB entries. The TLB units operated in serial order comparing the translation address with associated TLB entries until an identified one of the TLB units produces a hit. The TLB units following the TLB unit producing the hit might be disabled.

Abstract: The present invention provides a memory control device and a semiconductor processing apparatus which can be flexibly made adapted to a plurality of kinds of semiconductor memories. An SDRAM controller has: a register unit to which a command to be issued and a minimum interval (wait time) between issue of the command and issue of the next command are written by a CPU; and a command issuing unit that stops issue of the next command until the minimum interval written in the register elapses since issue of the command written in the register. Therefore, by changing software for the CPU, the SDRAM controller can be flexibly adapted to a plurality of kinds of SDRAMs.

Abstract: A non-volatile memory device is operated by outputting data in response to an alternating sequence of first and second edges of a read control signal, respectively. A determination is made whether the read control signal and a write control signal are in synchronization at one of the first edges. Output of the data is stopped at the second edge that follows the one of the first edges of the read control signal if the read control signal and the write control signal are in synchronization at the one of the first edges.

Abstract: A single data bus to a memory device can be split up into a number of data bus portions, each of which is managed by a different respective controller chip of multiple controller chips. During a memory access to a respective memory device, each of the multiple controller chips controls a different corresponding portion of the data bus to retrieve data from or store data to the memory device depending on whether the access is a read or write. To perform the data access, a synchronizer circuit (internal and/or external to the memory controller chips) synchronizes the multiple memory controller chips such that one of the memory controller chips drives the address bus and/or control signals to the memory device. After setting the address to the memory device, the memory controller chips either read data from or write data to the memory device based on the address.

Abstract: Disclosed is a method for reading data in a memory system including a buffer memory and a nonvolatile memory, the method being comprised of: determining whether an input address in a read request is allocated to the buffer memory; determining whether a size of requested data is larger than a reference unless the input address is allocated to the buffer memory; and conducting a prefetch reading operation from the nonvolatile memory if the requested data size is larger than the reference.

Abstract: Techniques for backing up distributed data are disclosed. In one particular exemplary embodiment, the techniques may be realized as a method for backing up distributed data comprising identifying one or more sources of distributed data targeted for backup, identifying two or more backup storage locations, determining which one or more backup storage locations of the two or more identified backup storage locations to utilize for a backup job based at least in part on one or more specified preferences, and configuring, for at least one of the sources of distributed data, the backup job using the one or more backup storage locations.

Abstract: An image compression and decompression method compresses data based upon the data states, and decompresses the compressed data based upon the codes generated during the compression.

Abstract: An I/O Optimizer receives an I/O request specifying a plurality of disk blocks of the disk drive for access. A plurality of I/O sub-requests is determined from the I/O request, each I/O sub-request specifying a set of one or more adjacent disk blocks of the plurality of disk blocks along the same cylinder. A plurality of execution sequences for performing the plurality of I/O sub-requests is determined. For each of the plurality of execution sequences, a total estimated execution time for performing the I/O sub-requests according to the execution sequence is calculated. One of the plurality of execution sequences for performing the I/O sub-requests is selected based, at least in part, on the total estimated execution times for the plurality of execution sequences. A disk drive controller is instructed to perform the I/O sub-requests according to the selected execution sequence.

Abstract: A concurrent and asynchronous system may be managed by monitoring the performance of a plurality of operations that access a designated region of memory. In that region of memory, an occurrence of a potentially non-deterministic event can be detected when at least one of the operations is a write operation. The occurrence of the potentially non-deterministic event may then be recorded.

Abstract: Disclosed are, inter alia, methods, apparatus, computer-readable media, mechanisms, and means for communicating with a memory device, such as by a memory controller, a refresh command at least partially overlapping in time with a read and/or write command. The refresh command typically specifies a group of locations (e.g., a bank) for being at least partially refreshed.

Abstract: A display control circuit capable of performing arbitration with the use of a simple configuration. The display control circuit exchanges, with a plurality of masters, attribute information defining conditions for displaying video on a display, and includes a memory for storing the attribute information, a plurality of channels associated with the respective masters for accepting, from the masters, access requests to access the memory, and an arbitration controller configured by hardware. The arbitration controller arbitrates the access requests accepted via the respective channels and permits a selected one of the access requests to access the memory.

Abstract: Concurrent threads in a multithreaded processor share access to a memory, with any location in the shared memory being accessible by any thread. In one embodiment, the shared memory has multiple independently-addressable memory banks, and one location per bank can be accessed in parallel. Parallel processing engines executing the threads generate a group of parallel memory access requests. Address conflict logic determines whether the requests can be satisfied in parallel (e.g., based on bank access constraints) and serializes the requests to the extent needed to avoid conflicts. In some embodiments, data read from one address in the shared memory can be broadcast to multiple processing engines.

Abstract: A shared memory system for a multicore computer system utilizing an interconnection network that furnishes tens of processing cores or more with the ability to refer concurrently to random addresses in a shared memory space with efficiency comparable to the typical efficiency achieved when referring to private memories. The network is essentially a lean and light-weight combinational circuit, although it may also contain non-deep pipelining. The network is generally composed of a sub-network for writing and a separate multicasting sub-network for reading, whose topologies are based on multiple logarithmic multistage networks, e.g. Baseline Networks, connected in parallel. The shared memory system computes paths between processing cores and memory banks anew at every clock cycle, without rearrangement.