Lsi Corporation

Seagate Technology Incorporated

Globalfoundries U.S. Inc.

Newisys, Inc.

International Business Machines Corporation

Samsung Electronics Co., Ltd.

Intel Corporation

Shashank Nemawarkar

Balakrishnan Sundararaman

David Sonnier

Shailendra Aulakh

David Brian Glasco

Rajesh Kota

Guru Prasadh

Carl Zeitler

Chris Randall Stone

Allen Vestal

Mark Ish

Bipul C Paul

Charles Edward Peet

Siddharth Krishna Kumar

William John Starke

Balaram Sinharoy

David S Levitan

Horia Cristian Simionescu

Navneet Jain

James David Dundas

Gregory Allen Northrop

Anant Baderdinni

Shivraj Gurpadappa Dharne

Madhu Saravana Sibi Govindan

Vivek Raj

Siddhartha Kumar Panda

Xiaoxin Fan

Larry Stephen King

Nishtha Gaul

Ming Yin

Bagavathy Raj Arunachalam

A disclosed non-volatile memory (NVM) structure is implemented in a fully depleted semiconductor-on-insulator technology processing platform and includes multiple NVM banks with NVM cells including transistors. NVM banks have well regions in a substrate. Transistors of NVM cells of each NVM bank are on an insulator layer above a corresponding well region for that bank. A bias control circuit causes well regions for NVM banks in a standby state to be biased with a reverse back biasing voltage and causes a well region for an NVM bank in an operational state to be biased with a forward back biasing voltage. The bias control circuit can initiate forward back biasing during a cache data retrieval process (before NVM bank access) to ensure that the corresponding well region of an NVM bank at issue is fully biased when, following the cache data retrieval process, access to the NVM bank is still required.

Hierarchical memory architecture including on-chip multi-bank non-volatile memory with low leakage and low latency

An apparatus and method for providing high throughput memory responses are provided. The apparatus includes a memory device including a plurality of memory arrays, a memory controller configured to control the memory device, the memory controller having a read queue, a write queue, and an address match circuit, and a data output circuit. The memory controller receives a read request, searches the write queue for a write address that matches a read address of the read request, and sends data associated with the write address from the write queue to the data output circuit without accessing the memory device when the write address matches the read address, the write address that matches the read address being a target address. The data output circuit outputs the data associated with the target address to an external device.

Apparatus and method for providing high throughput memory responses

The present disclosure relates to memory devices and, more particularly, to bias voltage generation circuit for memory devices and methods of operation. The voltage generation circuit includes: an internal voltage generator which providing a bias voltage to at least one internal node of a bias voltage generation circuitry; and at least one pre-charging circuitry providing a predefined bias voltage to at least one internal node including a distributed network of local drivers.

Bias voltage generation circuit for memory devices

Embodiments of the present disclosure provide a multi-port register file, including: a plurality of single-bit data registers for receiving and storing input data; a read path coupled to an output of each of the plurality of data registers; a plurality of AND gates, wherein an output of each of the plurality of data registers is coupled to an input of a respective AND gate of the plurality of AND gates; an input gating signal coupled to another input of each of the plurality of AND gates; a plurality of multi-bit registers, wherein an output of each of the plurality of AND gates is coupled to each of the plurality of multi-bit registers; and a write disable circuit coupled to the input gating signal for disabling a write signal applied to each of the plurality of multi-bit registers.

Multi-port register file for partial-sum accumulation

Embodiments of the present disclosure provide an apparatus including a memory array including a plurality of sub-arrays. A plurality of temporary storage units (TSUs) is coupled to the plurality of sub-arrays. Each TSU indicates whether the respective sub-array is undergoing one of a read operation and a write operation. A control circuit is coupled to each of the plurality of sub-arrays through a data bus. The control circuit transmits a read pulse or a write pulse as a first pulse with a delay in response to the sub-array undergoing the read operation or the write operation and transmits, instantaneously, the first pulse to one of the plurality of sub-arrays in response to the sub-array not undergoing the read operation or the write operation.

Apparatus and method for controlled transmitting of read pulse and write pulse in memory

Systems and methods are disclosed to perform a function level reset in a memory controller, in accordance with certain embodiments of the present disclosure. In some embodiments, an apparatus may comprise a storage controller circuit configured to receive a function reset indicator from a host device, the function reset indicator identifying a selected storage controller function executing at a storage controller of the apparatus. The circuit may abort each command associated with the selected function and pending at the apparatus based on the function reset indicator, verify that no commands associated with the selected function remain pending at the apparatus, and clear registers associated with the selected function based on the determination that no commands associated with the selected function remain.

Method and apparatus to perform a function level reset in a memory controller

According to one general aspect, an apparatus may include a main-branch target buffer (BTB). The apparatus may include a micro-BTB separate from and smaller than the main-BTB, and configured to produce prediction information associated with a branching instruction. The apparatus may include a micro-BTB confidence counter configured to measure a correctness of the prediction information produced by the micro-BTB. The apparatus may further include a micro-BTB misprediction rate counter configured to measure a rate of mispredictions produced by the micro-BTB. The apparatus may also include a micro-BTB enablement circuit configured to enable a usage of the micro-BTB's prediction information, based, at least in part, upon the values of the micro-BTB confidence counter and the micro-BTB misprediction rate counter.

Method and apparatus to control the use of hierarchical branch predictors based on the effectiveness of their results

Systems and methods are disclosed to abort a command at a data storage controller, in accordance with certain embodiments of the present disclosure. In some embodiments, an apparatus may comprise a data storage controller configured to receive an abort indicator from a host device, generate an abort tracking indicator at a receiving unit configured to receive commands from the host device, monitor to determine when the selected command is received at the receiving unit based on the abort tracking indicator, and abort the selected command when the selected command is received at the receiving unit. In some embodiments, the data storage controller may generate an abort tracking indicator at a completion unit configured to notify the host device of completed commands, and monitor for the selected command at the completion unit based on the abort tracking indicator.

Method and apparatus to abort a command

Method and apparatus for managing data in a distributed data storage system. In some embodiments, a plurality of storage devices define an overall available memory space. A control circuit stores a first copy of user data from a selected distributed data set in a working set of memory buffers, stores a duplicate, second copy of the user data in an alias set of memory buffers, generates parity data based on the second copy of the user data in the alias set of the memory buffers, and flushes the user data and the parity data from the alias set of memory buffers to the storage devices while the first copy of the user data remains in the working set of the memory buffers. In this way, subsequently received access commands can be serviced using the working set of the memory buffers.

Lockless parity management in a distributed data storage system

The disclosed technology provides for selection of a subset of available non-volatile memory devices in an array to receive a dirty cache data of a volatile cache responsive to detection of a power failure. In one implementation, the selection of the non-volatile memory devices is based on one or more predictive power parameters usable to estimate a time remaining during which a reserve power supply can support a cache offload to the selected subset of devices.

Cache offload based on predictive power parameter

Systems and methods are disclosed to delete a command queue, in accordance with certain embodiments of the present disclosure. An apparatus may comprise a circuit configured to receive a queue deletion indicator from a host device, including a queue identifier for a selected command queue to be deleted. The circuit may abort each command associated with the selected command queue and pending at the apparatus based on the queue identifier. Commands associated with the selected queue may be identified in a command table and flagged with an abort bit, which may signal an I/O processing pipeline to abort the command when encountered. The circuit may verify that no commands associated with the selected command queue remain pending at the apparatus, and send a completion indicator to notify the host device that the selected command queue is deleted.

Method and apparatus to delete a command queue

In response to a cacheable write request from a host, physical cache locations are allocated from a free list, and the data blocks are written to those cache locations without regard to whether any read requests to the corresponding logical addresses are pending. After the data has been written, and again without regard to whether any read requests are pending against the corresponding logical addresses, metadata is updated to associate the cache locations with the logical addresses. A count of data access requests pending against each cache location having valid data is maintained, and a cache location is only returned to the free list when the count indicates no data access requests are pending against the cache location.

Associative and atomic write-back caching system and method for storage subsystem

The disclosed technology provides for off-loading dirty data from a volatile cache memory to multiple non-volatile memory devices responsive to detection of a power failure. The arrangement of the dirty data is describable by a cache image, which is reconstructed within the volatile memory from the non-volatile memory devices responsive to detection of power restoration following the power failure.

Growing community of inventors

Austin, TX, United States of America

Shashank Nemawarkar