Lsi Corporation

Sandisk Technologies Inc.

Lsi Logic Corporation

Kioxia Corporation

Toshiba Memory Corporation

Senthil Thangaraj

Divya Reddy

Paresh Chatterjee

Basavaraj Gurupadappa Hallyal

Vinu Velayudhan

Sumant Kumar Patro

Girish Desai

Satish Babu Vasudeva

Rakesh Chandra

Narasimhulu Dharani Kotte

Barada Mishra

Steven E Wells

Aaron Keith Olbrich

Amit Jain

Parag Ranjan Maharana

Rodney Brittner

Mark Carlson

Varadaraj S Talamacki

Jimmy Sy

Venkatesh K Paulsamy

Gerald Edward Smith

James A Rizzo

Philip Rose

Saswati Das

Chayan Biswas

Ragendra Mishra

Ramya Subramanian

Gurpreet Singh Anand

Julien Magretts

In one embodiment, a solid state drive (SSD) comprises a plurality of non-volatile memory dies communicatively arranged in one or more communication channels, each of the plurality of non-volatile memory dies comprising a plurality of physical blocks, one or more channel controllers communicatively coupled to the one or more communication channels, respectively, and a memory controller communicatively coupled to the plurality of non-volatile memory dies via the one or more channel controllers, wherein the memory controller is configured to assign (i) the plurality of physical blocks of a first die of the plurality of non-volatile memory dies to only a first region and (ii) the plurality of physical blocks of a second die of the plurality of non-volatile memory dies to only a second region, perform only read operations on the first region in a first operation mode, and perform write operations or maintenance operations on the second region in a second operation mode concurrently with read operations on the first region in the first operation mode.

SSD architecture supporting low latency operation

In an SSD including a host interface, controller, non-volatile memory coupled to the controller, and volatile memory, data is stored in non-volatile memory in clusters having a cluster start address and a cluster end address. A mapping table is maintained in volatile memory, which indicates by a trim signature deallocated clusters in non-volatile memory. The trim table is populated with deallocated ranges and an associated offset within the non-volatile memory. The controller receives from the host interface a sequence of commands including a deallocate command including at least one address range including a start address and an end address, or a start address and a length of the address range. The controller processes the deallocate command by selecting one or both of a write-to-non-volatile-memory action and a record-to-volatile-memory action. In cases of power failure, the mapping table is restored for deallocated ranges at specific offsets recorded in the trim table.

Enhanced trim command support for solid state drives

Systems and methods disclosed herein allow for efficiently managing unmapped blocks to extend life of solid-state drives. In one aspect, a method includes: determining a quantity of unmapped storage units in the storage device and operating the storage device in a first mode of operation while the quantity satisfies a first threshold. The method also includes: changing to a second mode of operation if the quantity of unmapped storage units doesn't satisfy the first threshold. While operating in the second mode of operation, the method includes: determining whether processing a write command would reduce the quantity of unmapped storage units to a quantity less than a second threshold. If processing the write command wouldn't reduce the quantity of unmapped storage units to a quantity less than the second threshold, the method includes: accepting and processing the write command. Else, the method includes: forgoing acceptance and processing of the write command.

Efficiently managing unmapped blocks to extend life of solid state drive

Systems and methods disclosed herein allow for efficiently managing unmapped blocks to extend life of solid-state drives. In one aspect, a method includes: measuring a level of over-provisioning ('OP') in a storage device and operating it in a first mode of operation while the OP satisfies a first threshold. The method also includes: changing to a second mode of operation if the OP does not satisfy the first threshold. While operating in the second mode of operation, the method includes: (i) determining an unmapped portion of a declared storage capacity of the storage device; and (ii) determining whether processing a write command would reduce the unmapped portion to less than a second threshold. If processing the write command wouldn't reduce the unmapped portion to less than the second threshold, the method includes: accepting and processing the write command. Else, the method includes: forgoing acceptance and processing of the write command.

Efficiently managing unmapped blocks to extend life of solid state drive with low over-provisioning

Systems, methods, and/or devices are used to manage a storage system. In one aspect, the method includes, during a first time period: maintaining a credit pool for the first time period; limiting bandwidth used for transmitting data between a storage device of the storage system and a host operatively coupled with the storage device according to a status of the credit pool, where the storage device includes one or more memory devices; monitoring a temperature of the storage device; and, in accordance with a determination that a current temperature of the storage device exceeds a predetermined threshold temperature and the current temperature of the storage device satisfies one or more temperature criteria, reducing an initial value of the credit pool for a second time period according to a first adjustment factor corresponding to the predetermined temperature threshold, where the second time period is subsequent to the first time period.

Method and system for throttling bandwidth based on temperature

Systems, methods, and/or devices are used to manage a storage system. In one aspect, the method includes receiving, from a host to which a storage device of the storage system is operatively coupled, a request to perform a first memory operation on one or more memory devices of the storage device. The method includes determining a count of credits corresponding to the first memory operation. If a current count of credits in the first credit pool is greater than or equal to the count of credits corresponding to the first memory operation and a current count of credits in the second credit pool is greater than or equal to the count of credits corresponding to the first memory operation, the method includes: performing the first memory operation; and decrementing the first and second credit pools according to the count of credits corresponding to the first memory operation.

Method and system for throttling power consumption

A storage device and method enable processing of un-map commands. In one aspect, the method includes (1) determining whether a size of an un-map command satisfies (e.g., is greater than or equal to) a size threshold, (2) if the size of the un-map command satisfies the size threshold, performing one or more operations of a first un-map process, wherein the first un-map process forgoes (does not include) saving a mapping table to non-volatile memory of a storage device, and (3) if the size of the un-map command does not satisfy the size threshold, performing one or more operations of a second un-map process, wherein the second un-map process forgoes (does not include) saving the mapping table to non-volatile memory of the storage device and forgoes (does not include) flushing a write cache to non-volatile memory of the storage device.

Processing of un-map commands to enhance performance and endurance of a storage device

Methods and structure for task management in storage controllers of a clustered storage system. An initiator storage controller of the clustered storage system ships I/O requests for processing to a target storage controller of the system. Responsive to a need to abort a previously shipped I/O request, the initiator storage controller transmits a task management message to the target storage controller. The task management message identifies one or more previously shipped I/O requests to be aborted. The target storage controller processes the received task management message in due course of processing requests and completes processing for the aborted previously shipped request in an orderly manner. Resources associated with the aborted previously shipped requests are release within both controllers.

Methods and structure for task management in storage controllers of a clustered storage system

Disclosed is a virtual machine system where hardware timer interrupts are processed by a first virtual machine. The first virtual machine writes a timer value to a shared memory location while processing the hardware timer interrupt. The timer value may be based on a kernel timing parameter maintained by the operating system of the first virtual machine. A second virtual machine may read the shared timer value from the shared memory location in order to time inter-virtual machine processes such as I/O processing and I/O requests.

Inter-virtual machine time profiling of I/O transactions

Disclosed is a system with multiple virtual machines passing I/O requests via a shared memory space. A flag in shared memory is set to a first state in response to a first hypervisor I/O interrupt to indicate that an I/O processing routine is active (running). I/O requests are retrieved from an I/O queue in the shared memory by the I/O processing routine. Based on an indicator that there are no I/O requests remaining in said I/O queue, the shared flag is set to a second state to indicate that the I/O processing routine is deactivated (sleeping). In response to said shared flag being in the second state, when new I/O requests are going to be made, a second hypervisor I/O interrupt is generated. In response to said shared flag being in said first state, I/O requests are inserted into the I/O queue without generating a second hypervisor I/O interrupt.

Inter-virtual machine interrupt coalescing

The invention relates to a method for computer signal processing data and command transfer over an interface and more particularly to a communication between peripheral firmware and a host processor or Basic Input/Output System (BIOS) on a Peripheral Component Interconnect (PCI) bus. In one embodiment, a device and method for reducing the load on the PCI Bus is described. In yet another embodiment, a device and method is described for constructing a variable length command block comprising message frames and aligning all message frames for a particular command block that are contiguous in memory.

Variable length command pull with contiguous sequential layout

System and methods are disclosed for gathering debug information of a storage system of a computer system without requiring additional external hardware directly connected to the controller of the storage system.

Systems and methods for gathering debug information

A plurality of modes is provided for communicating between a host system and a peripheral storage system controller. A first communication mode may be selected from the plurality of communication modes based on a bit length required to communicate a physical address. During runtime, a switch from the first communication mode to a second communication mode may be performed in order to improve the efficiency of processing address requests at the storage system controller.

Address buffer mode switching for varying request sizes

Methods and systems for improved lock processing in a storage controller. Storage controllers that provide storage management features such as RAID storage controllers may concurrently process I/O requests received from attached host systems and I/O requests generated internally by the management processing in the controller. Such concurrent processing is coordinated by lock processing that allows affected storage areas to be locked during processing of an I/O operation. Features and aspects hereof allow such lock processing to lock regions flexibly defined by the controller. The flexible definition of the regions to be locked allows variance in the granularity of the locks required. Smaller granularity permits more concurrent I/O requests to be processed.

Growing community of inventors

Fremont, CA, United States of America

Senthil Thangaraj