Abstract: Architectures and methods for performing big data analytics by providing an integrated storage/processing system containing non-volatile memory devices that form a large, non-volatile memory array and a graphics processing unit (GPU) configured for general purpose (GPGPU) computing. The non-volatile memory array is directly functionally coupled (local) with the GPU and optionally mounted on the same board (on-board) as the GPU.

Abstract: A modular mass storage system and method that enables cableless mounting of ATA and/or similar high speed interface-based mass storage devices in a computer system. The system includes a printed circuit board, a system expansion slot interface on the printed circuit board and comprising power and data pins, a host bus controller on the printed circuit board and electrically connected to the system expansion slot interface, docking connectors connected with the host bus controller to receive power and exchange data therewith and adapted to electrically couple with industry-standard non-volatile memory devices without cabling therebetween, and features on the printed circuit board for securing the memory devices thereto once coupled to the docking connectors.

Abstract: Memory technology adapted to store data in a binary format. Such technology includes a semiconductor memory device having memory cells, each having a substrate and at least three graphene layers that are oriented to define a graphene stack disposed in a plane. The graphene stack of each memory cell is connected to a bit line and to a ground connection so that a conductive path is defined in the plane of the graphene stack. The in-plane conductivity of the graphene stack of each memory cell is altered during programming of the memory cell to define a binary value of bits stored in the memory cell.

Abstract: RAID storage systems and methods adapted to enable the use of NAND flash-based solid-state drives. The RAID storage system includes an array of solid-state drives and a controller operating to combine the solid-state drives into a logical unit. The controller utilizes data striping to form data stripe sets comprising data (stripe) blocks that are written to individual drives of the array, utilizes distributed parity to write parity data of the data stripe sets to individual drives of the array, and writes the data blocks and the parity data to different individual drives of the array. The RAID storage system detects the number of data blocks of at least one of the data stripe sets and then, depending on the number of data blocks detected, may invert bit values of the parity data or add a dummy data value of “1” to the parity value.

Abstract: A mass storage system comprising multiple memory cards, each with non-volatile memory components, a system bus interface for communicating with a system bus of a host system, and at least one ancillary interface. The ancillary interface is configured for direct communication of commands, addresses and data between the memory cards via a cross-link connector without accessing the system bus interface.

Abstract: A resistive random access memory integrated circuit for use as a mass storage media and adapted for bulk erase by substantially simultaneously switching all memory cells to one of at least two possible resistive states. Bulk switching is accomplished by biasing all bottom electrodes within an erase area to a voltage lower than that of the top electrodes, wherein the erase area can comprise the entire memory array of the integrated circuit or else a partial array. Alternatively the erase area may be a single row and, upon receiving the erase command, the row address is advanced automatically and the erase step is repeated until the entire array has been erased.

Abstract: A test-ahead feature for non-volatile memory-based mass storage devices to anticipate device failure. The test-ahead feature includes a method performed with a solid-state mass storage device having a controller, a cache memory, and at least one non-volatile memory device. At least a first block is reserved on the at least one non-volatile memory device as a wear-indicator block and a plurality of second blocks are used for data storage. Information is stored corresponding to the number of write and erase cycles encountered by the second blocks during usage of the mass storage device, and the information is accessed to perform wear leveling among the second blocks. The wear-indicator blocks are subjected to an offset number of write and erase cycles in excess of the number of write and erase cycles encountered by the second blocks, after which an integrity check of the first block is performed.

Abstract: RAID storage systems and methods adapted to enable the use of NAND flash-based solid-state drives. The RAID storage system includes an array of solid-state drives and a controller operating to combine the solid-state drives into a logical unit. The controller utilizes data striping to form data stripe sets comprising data (stripe) blocks that are written to individual drives of the array, utilizes distributed parity to write parity data of the data stripe sets to individual drives of the array, and writes the data blocks and the parity data to different individual drives of the array. The RAID storage system detects the number of data blocks of at least one of the data stripe sets and then, depending on the number of data blocks detected, may invert bit values of the parity data or add a dummy data value of “1” to the parity value.

Abstract: A solid-state mass storage device and method of operating the storage device to anticipate the failure of at least one memory device thereof before a write endurance limitation is reached. The method includes assigning at least a first memory block of the memory device as a wear indicator that is excluded from use as data storage, using pages of at least a set of memory blocks of the memory device for data storage, writing data to and erasing data from each memory block of the set in program/erase (P/E) cycles, performing wear leveling on the set of memory blocks, subjecting the wear indicator to more P/E cycles than the set of memory blocks, performing integrity checks of the wear indicator and monitoring its bit error rate, and taking corrective action if the bit error rate increases.

Abstract: Methods of operating a non-volatile solid state memory-based mass storage device having at least one non-volatile memory component. In one aspect of the invention, the one or more memory components define a memory space partitioned into user memory and over-provisioning pools based on a P/E cycle count stored in a block information record. The storage device transfers the P/E cycle count of erased blocks to a host and the host stores the P/E cycle count in a content addressable memory. During a host write to the storage device, the host issues a low P/E cycle count number as a primary address to the content addressable memory, which returns available block addresses of blocks within the over-provisioning pool as a first dimension in a multidimensional address space. Changed files are preferably updated in append mode and the previous version can be maintained for version control.

Abstract: Solid-state mass storage devices, host computer systems, and methods of managing non-volatile solid-state memory components used therein. The memory components comprise memory cells organized in functional units that are adapted to receive units of data transferred from the host computer system and correspond to the functional units of the memory component. The level of programming for each cell is reduced by performing an analysis of the bit values of the units of data to be written to at least a first of the functional units of the memory component. Depending on the analysis of “0” and “1” bit values of the units of data to be written, the bit values are inverted before writing the units of data to the first memory component.

Abstract: A solid-state mass storage device adapted to be used as a cache for an hard disk drive that utilizes a more efficient logical data management method relative to conventional systems. The storage device includes a circuit board, a memory controller, at least one non-volatile memory device, and at least two data interfaces. The storage device is coupled to a host computer system and configured to operate as a cache for at least one hard disk drive. The storage device is interposed between the host computer system and the at least one hard disk drive. Both the storage device and the at least one hard disk drive are coupled to the host computer system through a single connection and configured to operate in a daisy chain configuration.

Abstract: Methods and systems for mass storage of data over two or more tiers of mass storage media that include nonvolatile solid-state memory devices, hard disk devices, and optionally volatile memory devices or nonvolatile MRAM in an SDRAM configuration. The mass storage media interface with a host through one or more PCIe lanes on a single printed circuit board.

Abstract: Architectures and methods for performing big data analytics by providing an integrated storage/processing system containing non-volatile memory devices that form a large, non-volatile memory array and a graphics processing unit (GPU) configured for general purpose (GPGPU) computing. The non-volatile memory array is directly functionally coupled (local) with the GPU and optionally mounted on the same board (on-board) as the GPU.

Abstract: Computer systems and methods that utilize a GPU whose operation is able to switch between ECC and non-ECC memory operations on demand. The computer system includes a graphics processing unit and a memory controller and local memory that are functionally integrated with the graphics processing unit. The memory controller has at least two operating modes comprising a first memory access mode that uses error checking and correction when accessing the local memory, and a second memory access mode that does not use error checking and correction when accessing the local memory. The memory controller is further operable to switch the operation of the memory controller between the first and second memory access modes without rebooting the computer system.

Abstract: A non-volatile solid state memory-based mass storage device having at least one non-volatile memory component and methods of operating the storage device. In one aspect of the invention, the one or more memory components define a memory space partitioned into user memory and over-provisioning pools based on a P/E cycle count stored in a block information record. The storage device transfers the P/E cycle count of erased blocks to a host and the host stores the P/E cycle count in a content addressable memory. During a host write to the storage device, the host issues a low P/E cycle count number as a primary address to the content addressable memory, which returns available block addresses of blocks within the over-provisioning pool as a first dimension in a multidimensional address space. Changed files are preferably updated in append mode and the previous version can be maintained for version control.

Abstract: A mass storage device and method that utilize storage memory and a shadow memory capable of increasing the speed associated with copying data from one location to another location within the storage memory without the need to access a host computer for the copy transaction. A controller of the mass storage device receives a file copy request for a file to be copied between first and second locations within the storage memory. Data from the first location within the storage memory is then loaded into a shadow memory means of the mass storage device, and then the data is written from the shadow memory means to the second location within the storage memory.

Abstract: A host server computer system that includes a hypervisor within a virtual space architecture running at least one virtualization, acceleration and management server and at least one virtual machine, at least one virtual disk that is read from and written to by the virtual machine, a cache agent residing in the virtual machine, wherein the cache agent intercepts read or write commands made by the virtual machine to the virtual disk, and a solid state drive. The solid state drive includes a non-volatile memory storage device, a cache device and a memory device driver providing a cache primitives application programming interface to the cache agent and a control interface to the virtualization, acceleration and management server.

Abstract: A cooling system and method for cooling electronic components, including IC dies. The cooling system employs a cooling device that includes a composite structure having first and second plates arranged substantially in parallel and bonded together to define a sealed cavity therebetween. The first plate has a surface that defines an outer surface of the composite structure and is adapted for thermal contact with at least one electronic component. A mesh of interwoven strands is disposed within the cavity and lies in a plane substantially parallel to the first and second plates, with the strands bonded to the first and second plates. A fluid is contained and sealed within the cavity of the composite structure, and flows through interstices defined by and between the strands of the mesh.

Abstract: A solid-state mass storage device and method of operating the storage device to anticipate the failure of at least one memory device thereof before a write endurance limitation is reached. The method includes assigning at least a first memory block of the memory device as a wear indicator that is excluded from use as data storage, using pages of at least a set of memory blocks of the memory device for data storage, writing data to and erasing data from each memory block of the set in program/erase (P/E) cycles, performing wear leveling on the set of memory blocks, subjecting the wear indicator to more P/E cycles than the set of memory blocks, performing integrity checks of the wear indicator and monitoring its bit error rate, and taking corrective action if the bit error rate increases.