摘要:
First data is received for storing in a first asymmetric memory device. A first writing phase is identified as a current writing phase. A first segment included in the first asymmetric memory device is identified as next segment available for writing data. The first data is written to the first segment. Information associated with the first segment is stored, along with information indicating that the first segment is written in the first writing phase. Second data is received for storing in the asymmetric memory. A second segment included in the first asymmetric memory device is identified as the next segment available for writing data. The second data is written to the second segment. Information associated with the second segment and the second memory block is stored along with information indicating that the second segment is written in the second writing phase.
摘要:
Some implementations include a method of managing a hosted non-volatile random-access memory (NVRAM) based storage subsystem that includes NVRAM devices. The method includes: receiving, at a device driver on the host computing device, write requests each requesting to write a respective unit of data to the NVRAM-based storage subsystem; categorizing the write requests into subgroups of write requests, where write requests within respective subgroups are mutually exclusive; ascertaining a load condition of each of several of the NVRAM devices of the NVRAM-based storage subsystem; identifying a target location on at least one NVRAM device to service a particular subgroup of write requests according to the ascertained load conditions of the NVRAM devices of the NVRAM-based storage subsystem; and servicing the particular subgroup of write requests by writing the corresponding units of data to the identified target location on the at least one NVRAM device of the NVRAM-based storage subsystem.
摘要:
In one embodiment of the invention, a memory system includes non-volatile-memory-devices (NVMDs) coupled to memory channels to share busses and a memory controller coupled to the memory channels in communication between the plurality of NVMDs. Each NVMD independently executes a read, write, or erase operation at a time. The memory controller includes channel schedulers to schedule control and data transfers associated with the read, write, and erase operations on the memory channels; and high priority and low priority queues coupled to the channel schedulers. The channel schedulers prioritize operations waiting in the high priority queues over operations waiting in the low priority queues. The channel schedulers further prioritize read operations waiting in either the high priority queue or the low priority queue over write and erase operations waiting in each respective queue.
摘要:
In one embodiment of the invention, a memory system includes non-volatile-memory-devices (NVMDs) coupled to memory channels to share busses and a memory controller coupled to the memory channels in communication between the plurality of NVMDs. Each NVMD independently executes a read, write, or erase operation at a time. The memory controller includes channel schedulers to schedule control and data transfers associated with the read, write, and erase operations on the memory channels; and high priority and low priority queues coupled to the channel schedulers. The channel schedulers prioritize operations waiting in the high priority queues over operations waiting in the low priority queues. The channel schedulers further prioritize read operations waiting in either the high priority queue or the low priority queue over write and erase operations waiting in each respective queue.
摘要:
In one embodiment of the invention, a system is disclosed including a master memory controller and a plurality of memory modules coupled to the master memory controller. Each memory module includes a plurality of read-writeable non-volatile memory devices in a plurality of memory slices to form a two-dimensional array of memory. Each memory slice in each memory module includes a slave memory controller coupled to the master memory controller. When the master memory controller issues a memory module request, it is partitioned into a slice request for each memory slice.
摘要:
In one embodiment of the invention, a memory apparatus for improved write performance is disclosed. The memory apparatus includes a base printed circuit board (PCB) having an edge connector for plugging into a host server system; a card level power source to provide card level power during a power failure; a memory controller coupled to the card level power source and having one or more memory channels; and one or more non-volatile memory devices (NVMDs) coupled to the card level power source and organized to respectively couple to the memory channels controlled by the memory controller. Each memory controller provides queuing and scheduling of memory operations on a channel for each NVMD in the memory channels. Responsive to power failure, the memory controller receives card level power and changes the scheduling of memory operations to the NVMDs in each memory channel.
摘要:
In one embodiment of the invention, a memory apparatus for improved write performance is disclosed. The memory apparatus includes a base printed circuit board (PCB) having an edge connector for plugging into a host server system; a card level power source to provide card level power during a power failure; a memory controller coupled to the card level power source and having one or more memory channels; and one or more non-volatile memory devices (NVMDs) coupled to the card level power source and organized to respectively couple to the memory channels controlled by the memory controller. Each memory controller provides queuing and scheduling of memory operations on a channel for each NVMD in the memory channels. Responsive to power failure, the memory controller receives card level power and changes the scheduling of memory operations to the NVMDs in each memory channel.
摘要:
A memory controller (MC) is associated with a remapping table to enable access to content in a memory system that includes asymmetric memory. The MC receives a request for a memory read or an Input/Output (I/O) write from a central processing unit (CPU) for a physical address specified by the system's memory management unit (MMU). The CPU uses the MMU to manage memory operations for the CPU, by translating the virtual addresses associated with CPU instructions into physical addresses representing system memory or I/O locations. The MC for asymmetric memories is configured to process the MMU-specified physical addresses as an additional type of virtual addresses, creating a layer of abstraction between the physical address specified by the MMU and the physical memory address with which that address is associated by the MC. The MC shields the CPU from the computational complexities required to implement a memory system with asymmetric components.
摘要:
A computing system is disclosed that includes a memory controller in a processor socket normally reserved for a processor. A plurality of non-volatile memory modules may be plugged into memory sockets normally reserved for DRAM memory modules. The non-volatile memory modules may be accessed using a data communication protocol to access the non-volatile memory modules. The memory controller controls read and write accesses to the non-volatile memory modules. The memory sockets are coupled to the processor socket by printed circuit board traces.The data communication protocol to access the non-volatile memory modules is communicated over the printed circuit board traces and through the sockets normally used to access DRAM type memory modules.
摘要:
A memory controller writes to a virtual address associated with data residing within an asymmetric memory component of main memory that is within a computer system and that has a symmetric memory component, while preserving proximate other data residing within the asymmetric memory component. The symmetric memory component within the main memory of the computer system is configured to enable random access write operations in which an address within a block of the symmetric memory component is written without affecting the availability of other addresses within the block of the symmetric memory component during the writing of that address. The asymmetric memory component is configured to enable block write operations in which writing to an address within a region of the asymmetric memory component affects the availability of other addresses within the region of the asymmetric memory component during the block write operations involving the address.