Memory device, memory controller and memory system
By generating concurrent refresh and pre-charge commands in the memory controller, grouping operations across multiple memory banks, the problem of difficult scaling of traditional memory bank architectures is solved, the performance and bandwidth utilization of memory is improved, and power consumption and chips are reduced. size.
Patent Information
- Application Number
- CN202011307668.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2015-09-25
- Filing Date
- 2016-02-12
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2036-02-12
AI Technical Summary
Traditional memory bank architectures are difficult to scale when the memory subsystem scale and speed increase, resulting in the core cycle time being unable to scale with the data I/O bandwidth. Increasing the core cycle time will lead to power concentration and chip size increase. The memory bank packet architecture divides memory resources. Access is handled independently but the operation bandwidth of refresh commands is increased.
By generating concurrent refresh and pre-charge commands in the memory controller, using a single command to operate multiple memory banks across multiple memory banks, the number of refresh commands issued to the memory controller is reduced and the circulation of memory banks is improved Time and data bandwidth utilization.
It realizes more efficient memory bank refresh and pre-charge in the memory bank packet architecture, reduces the system's power consumption and chip size, and improves the memory performance and bandwidth utilization.
Smart Images

Figure CN112382322B_ABST
Abstract
Description
[0001] This application is a divisional application, the application number of the parent application is 2016800077119, the application date is February 12, 2016, and the name of the invention is "Precharging and refreshing a memory cell in a memory device having a memory cell grouping architecture".
[0002] Related Applications
[0003] This application is a non-provisional application based on U.S. Provisional Application No. 62 / 126,597 filed on February 28, 2015. This application claims the benefit of priority to the provisional application, and the provisional application is hereby incorporated herein by reference. Technical Field
[0004] Embodiments of the present invention relate generally to memory subsystems, and more particularly to performing refresh in a memory device having a bank group architecture.
[0005] Copyright Notice / License
[0006] Portions of the disclosure of this patent document may contain material that is protected by copyright. The copyright owner has no objection to the reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever. This copyright notice applies to all data described below and in the accompanying drawings, and to any software described below: 2015, Intel Corporation, All rights reserved. Background Art
[0007] Some memory devices have a bank grouping architecture. Traditional bank architectures do not scale well as the size and speed of memory subsystems increase. Improvements in the technology used to produce newer memories tend to make the features available in the memory smaller, which can reduce operating voltages and improve speed. However, as processing technology improves, the size and complexity of memories increase, and therefore the core cycle time (tCCD) does not scale with the data I / O (input / output) bandwidth. tCCD indicates a column-to-column command or command-to-command delay, indicating two consecutive (back-to-back) accesses to the same memory bank. A column command can indicate a column address strobe command (CAS) for accessing memory.
[0008] As I / O speeds increase, the core traditionally needs to increase its speed to utilize the full data I / O bandwidth. Therefore, doubling the I / O speed requires doubling the core cycle time. However, increasing the core cycle time to reduce tCCD is very power-intensive and die-size-intensive. Utilizing a bank grouping architecture, memory resources can be segmented into groups of banks. Using bank groupings can improve the apparent cycle time of the memory because access to different bank groups can be handled independently. Therefore, instead of needing to cycle quickly twice to fully utilize the I / O bandwidth, the memory can access different bank groups sequentially. The memory subsystem can allow shorter cycles for accesses between different bank groups, as well as longer cycles for accesses between banks within the same bank grouping.
[0009] Although the use of bank grouping provides an improvement in data bandwidth utilization, this can significantly increase command bandwidth. The use of bank grouping typically increases the number of banks in the system. With more banks, the memory controller needs to increase the number of commands for maintaining the memory. In particular, in volatile memory devices, refresh commands consume a significant amount of operating bandwidth of the memory by accessing banks in different bank groupings. Increasing the number of banks typically increases the number of refresh commands that the memory controller needs to issue. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] The following description includes a discussion of the accompanying drawings with examples illustrated by way of examples of implementations of embodiments of the present invention. The accompanying drawings should be understood by way of example rather than by way of limitation. As used herein, reference to one or more "embodiments" will be understood to describe specific features, structures, and / or characteristics included in at least one implementation of the present invention. Therefore, phrases such as "in one embodiment" or "in an alternate embodiment" that appear herein describe various embodiments and implementations of the present invention, and do not necessarily all refer to the same embodiment. However, they are not necessarily mutually exclusive.
[0011] Figure 1 is a block diagram of an embodiment of a system that implements bank refresh across bank groups.
[0012] Figure 2 is a representation of an embodiment of a command table for per bank refresh and all bank refresh commands and per bank precharge and all bank precharge commands.
[0013] Figure 3 is a graphical representation of an embodiment of system performance of a system implementing bank refresh across bank groups.
[0014] Figure 4is a flow diagram of an embodiment of a process for performing a refresh.
[0015] Figure 5 is a block diagram of an embodiment of a computing system in which refreshing of designated memory banks across memory bank groups can be implemented.
[0016] Figure 6 is a block diagram of an embodiment of a mobile device in which refreshing of a designated memory bank across memory bank groups can be implemented.
[0017] A description of specific details and implementations follows, including descriptions of figures that may depict some or all of the embodiments described below, as well as a discussion of other potential embodiments or implementations of the inventive concepts presented herein. DETAILED DESCRIPTION
[0018] As described herein, memory subsystem refresh management enables the use of a single refresh command to refresh one or more identified memory banks across different memory bank groupings. Similar capabilities can exist in terms of implementing the use of a single precharge command to precharge one or more identified memory banks across different memory bank groupings. Instead of sending precharge commands and / or refresh commands identifying a memory bank or multiple memory banks in a separate memory bank grouping individually by each memory bank grouping, the memory controller can generate commands that cause the memory device to concurrently precharge and / or refresh memory banks in different memory bank groupings.
[0019] References to memory devices can apply to different memory types, and in particular any memory with a bank grouping architecture. Memory devices generally refer to volatile memory technology. Volatile memory is a memory whose state (and therefore the data stored thereon) is uncertain if power to the device is interrupted. Non-volatile memory refers to a memory whose state is determined even if power to the device is interrupted. Dynamic volatile memory requires refreshing of the data stored in the device to maintain state. An example of dynamic volatile memory includes DRAM (dynamic random access memory), or some variants such as synchronous DRAM (SDRAM). The memory subsystem described herein may be compatible with a variety of memory technologies, such as DDR4 (DDR version 4, initial specification released by JEDEC in September 2012), DDR4E (under development by JEDEC), LPDDR4 (Low Power Double Data Rate (LPDDR) version 4, JESD209-4, initially released by JEDEC in August 2014), WIO2 (Wide I / O 2 (WideIO2), JESD229-2, initially released by JEDEC in August 2014), HBM (High Bandwidth Memory DRAM, JESD235, initially released by JEDEC in October 2013), DDR5 (DDR version 5, currently under discussion by JEDEC), LPDDR5 (currently under discussion by JEDEC), HBM2 (HBM version 2, currently under discussion by JEDEC), and / or other, as well as derivative or extended technologies based on these specifications.
[0020] In one embodiment, in addition to or in lieu of volatile memory, for a device having a bank grouping architecture, a reference to a memory device may indicate a non-volatile memory whose state is determined even if power to the device is interrupted. In one embodiment, the non-volatile memory device is a block addressable memory device, such as NAND or NOR technology. Therefore, the memory device can also include another generation of non-volatile devices, such as a three-dimensional cross-point memory device, or other byte addressable non-volatile memory devices. In one embodiment, the memory device can be or include a multi-threshold level NAND flash memory, a NOR flash memory, a single-level or multi-level phase change memory (PCM), a resistive memory, a nanowire memory, a ferroelectric transistor random access memory (FeTRAM), a magnetoresistive random access memory (MRAM) memory containing a memristor technology, or a spin transfer torque (STT)-MRAM, or a combination of any of the above memories, or other memories.
[0021] More specifically, the subject matter described herein applies to memory technologies that support and / or are compatible with the use of memory bank groupings. Certain access commands addressed to a particular memory bank can be executed across all memory bank groups at the identified memory bank address.
[0022] Figure 1 1 is a block diagram of an embodiment of a system for implementing memory bank refresh across memory bank groups. System 100 is an example of a memory subsystem having a memory controller 110 and a memory device 120. Memory device 120 can be or include, for example, a DRAM chip or die. In one embodiment, memory device 120 is an SDRAM chip or die, or includes an SDRAM chip or die. In one embodiment, memory 120 is an independent memory device in the system. In one embodiment, memory 120 represents one or more memory devices directly connected to a memory controller and / or a host processor. In one embodiment, memory 120 represents a plurality of memory devices coupled in parallel to a controller and / or a host processor that shares one or more buses (e.g., a command and address bus and at least a portion of a data bus). In one embodiment, memory 120 is one of a plurality of memory devices that are part of a memory module (e.g., a dual in-line memory module (DIMM)).
[0023] The memory controller 110 includes command logic 112 for managing the sending of commands to the memory 120. The command logic 112 represents logic for determining what commands to issue, determining when to issue commands, and generating commands to be sent to the memory devices. In one embodiment, the command logic 112 is part of the other logic or control logic of the memory controller 110. The memory controller 110 includes I / O (input / output) hardware 114 for connecting to the memory 120. The I / O hardware 114 can include pins, connectors, traces, wires, signal lines, and / or other hardware resources for exchanging signals between the memory controller 110 and the memory 120. The I / O hardware 114 can include drivers, transceivers, ODTs (on-chip termination resistors), and / or other logic for sending and / or receiving signals.
[0024] In one embodiment, the command logic 112 includes refresh command logic 142 for managing the refresh of the volatile memory 120. The refresh logic 142 represents logic that determines when to issue a refresh command and generates commands to be sent to the memory device. In one embodiment, the command logic 112 includes precharge command logic 144 for managing the precharge of portions of the memory array or memory resources 130 of the memory device 120. In one embodiment, the memory controller 110 generates one or more precharge commands in conjunction with one or more refresh commands. The precharge logic 144 enables the memory 120 to place the memory resources in a reset state.
[0025] The memory 120 includes a memory resource 130 organized into a plurality of memory bank groups 132. Thus, the memory 120 has a memory bank grouping architecture. For example, the memory 120 can include 2, 4, or 8 memory bank groups 132. Each memory bank grouping includes a plurality of memory banks 134, for example, 4 or 8 memory banks. For example, the memory 120 can be an x4 or x8 device (reference data bus interface width) having 4 memory bank groups, each memory bank grouping having 8 memory banks, or an x4 or x8 device having 8 memory bank groups, each memory bank grouping having 4 memory banks, or an x16 device having 4 memory bank groups, each memory bank grouping having 4 memory banks. These examples are for purposes of illustration and not limitation. Each memory bank includes an addressable memory location (e.g., a word line or a cache line or a page) for storing data. In a specific example, the memory 120 is a DDR4 or DDR4E memory having a memory bank grouping architecture. As mentioned above, due to the core cycle time (tCCD), the traditional memory bank architecture does not scale. However, some DDR4 architectures can have 2 or 4 memory bank groups, each with up to eight memory banks. DDR4 can allow a short core cycle delay time tCCD_S (which is 4 cycles) for accessing different memory bank groups, and a long core cycle delay time tCCD_L (which is greater than 4 cycles) for accessing memory banks within the same memory bank group. tCCD_S can be set equal to half the burst length (BL / 2) to allow data to be streamed at full bandwidth for reading and writing. Therefore, a burst length of 8 corresponds to a tCCD_S of 4 cycles.
[0026] The memory 120 includes access logic 122 for receiving and decoding commands from the memory controller 110. In one embodiment, the access logic 122 is a part of a controller or control logic on a memory device that manages the operation of the memory. Such a controller refers to an on-memory controller that controls the internal timing and sequencing of operations to execute commands received from the memory controller 110. The internal controller generates internal operations to execute the received commands. The memory 120 includes an I / O 124 for interfacing with the memory controller 110. The I / O 124 can be similar or identical to the I / O 114 of the memory controller, and can likewise represent any type of hardware that exchanges signals between the two devices.
[0027] In one embodiment, the access logic 122 includes decode logic 150. The decode logic 150 enables the memory 120 to decode commands received from the memory controller 110 to determine what type of access is to be made to which memory resources 130 of the memory. The decode logic 150 enables the memory 120 to route the command signal to the memory resource 130 identified in the command. In one embodiment, the access logic 122 includes refresh logic 152 internal to the memory 120. In one embodiment, the access logic 122 includes precharge logic 154 internal to the memory 120.
[0028] In one embodiment, the memory controller 110 can issue a refresh command (REF) and a refresh command per bank (REFpb). The REF command triggers the memory 120 to refresh the value stored in the memory location associated with the address information in the REF command via the refresh logic 152. The REFpb command operation is similar to the REF command, but specifically performs the operation of refreshing the memory bank. Therefore, in the memory 120 with a bank grouping architecture, the memory controller can be enabled to control access by interleaving the memory banks. The memory 120 can be configured to implement commands that are interleaved across banks. The timing of accessing across banks can be faster than the timing of accessing the same bank, allowing the memory controller to more accurately control the access timing of issued commands.
[0029] With each bank refresh command (REFpb), the memory controller 110 is able to direct the refresh at a specific bank. For example, in refresh management for DDR4E devices, address bit A12 is used to distinguish between a refresh command (REF) and a refresh command for each bank (REFpb). A refresh command with bit A12=1 indicates that all banks are refreshed, while a refresh command with A12=0 indicates that the specified (multiple) banks are refreshed. Therefore, the memory controller 110 can generate and send refresh commands for refreshing all banks (REF) and / or specific banks (REFpb). For example, in a refresh command, the values on bank address bits BA2, BA1, and BA0 specify which bank is to be refreshed. In one embodiment, the bank identified by bank address bits BA2, BA1, and BA0 is the bank to be refreshed in all bank groups when the REFpb command is issued. Thus, in one embodiment, the Refresh Each Bank command identifies the banks to be refreshed and the memory device 120 will refresh the identified banks across all bank groups 132 (eg, for BA[2:0]=000, bank '000' in BankGroup0, BankGroup1, etc. will be refreshed).
[0030] In one embodiment, the memory controller 110 can issue a precharge command, which is similarly defined as precharging the same memory bank in all memory bank groups. For example, the system can define up to three different types of precharge commands. In one embodiment, a precharge command (PRE) with bit A10=0 or L (low) and A12=1 or H (high) can mean precharging a single memory bank indicated by the memory bank grouping and memory bank specified by BG1, BG0 and BA2, BA1, BA0. In one embodiment, a precharge command (PREpb) with A10=0 and A12=0 can mean precharging each memory bank, precharging the same memory bank in all memory bank groups as the memory bank specified by BA2, BA1, and BA0. In one embodiment, a precharge command (PREA) with A10=1 can mean precharging all memory banks. Via precharge logic 154, the memory 120 can implement a precharge command received from the memory controller 110. Thus, the PREA command can set all banks to an idle state, the PRE can set a specific bank in a specific bank group to an idle state, and the PREPb command can set a specified bank to an idle state across all bank groups. In one embodiment, the bank identified by the address bits BA[2:0] is the bank to be precharged in all bank groups when PREPb is issued. In one embodiment, the memory controller 110 issues a corresponding precharge command with a refresh command (e.g., PREA is issued with REF, and PREPb is issued with REFpb).
[0031] Conventionally, each REFpb command increments an internal bank counter. The memory controller can compare the internal bank counter to a threshold representing the number of available banks in bank grouping 132. In one embodiment, upon detecting that the internal bank counter is equal to the number of available banks in bank grouping 132, the memory controller 110 can reset the counter to restart on the next REFpb command or a subsequent REFpb command. Each time the internal bank counter is reset and restarted on the next REFpb command, the memory controller 110 also increments the global refresh counter.
[0032] In one embodiment, the memory controller 110 can issue REFpb commands to any memory bank 134 in any memory bank order, as long as each memory bank 134 has a REFpb command issued to it before issuing a subsequent REFpb command to any memory bank. In one embodiment, the memory controller 110 issues a corresponding PRE command with each REFpb command, and can therefore send precharge commands and refresh commands in the same memory bank order. The memory bank order refers to the sequence of refreshes. The round robin order can be refreshed in the order of the memory bank addresses, for example, from highest to lowest or from lowest to highest. The memory controller can interleave refreshes and or precharges to access the memory banks out of order relative to the memory bank addresses. The memory controller 110 can manage the memory bank order and therefore prevent the issuance of subsequent REFpb commands issued to the same memory bank before each memory bank receives the REFpb command. The memory controller can similarly manage PRE commands.
[0033] In one embodiment, issuing a REFpb command multiple times to the same bank before issuing REFpb to all banks in a bank group is not permitted. In one embodiment, the first REFpb command issued is a "sync" REFpb command, and the "sync" count resets the internal bank counter to zero when any of the following occurs: (a) each bank has received a REFpb command; (b) the memory controller applies RESET; (c) the memory device enters or exits self-refresh mode; or (d) the memory controller issues a REF command to refresh all banks (i.e., a general refresh command, not a REFpb command).
[0034] In one embodiment, the memory controller 110 includes counter logic 116. Counter logic 116 can represent one or more counters internal to the memory controller 110. Counters can be used to manage the issuance of commands (e.g., refresh commands). The memory controller 110 can use one or more counters 116 to manage the refreshing of rows in the memory 120. With each refresh, the memory controller 110 can increment the counters 116 to track which rows have been refreshed. In one embodiment, the counters 116 can include bank counters and global counters.
[0035] In one embodiment, memory 120 includes counter 126, which represents one or more counters within a memory device. In one embodiment, memory 120 uses counter 126 to keep track of row refreshes. In one embodiment, counter 116 can include a bank counter and a global counter. In one embodiment, system 100 includes either counter 116 or counter 126, but not both. In response to a refresh command received from memory controller 110, memory 120 can manage counter 126 using internal control logic (internal controller).
[0036] In one embodiment, counter 126 represents a global refresh counter within memory 120. In one embodiment, global refresh counter 126 can be incremented when either the memory controller issues a REF command to refresh all banks, or all banks have received a REFpb command and the "sync" count is reset to zero. A REF command issued when the bank counter is not zero will increment the global refresh counter of the memory device. In one embodiment, a REFpb command cannot be issued to a memory device until the following conditions are met:
[0037] - After the previous REF command, tRFC (refresh cycle time) has been met;
[0038] - After the previous REFpb command, tRFCpb (each bank refresh cycle time) has been satisfied;
[0039] - tRP (precharge cycle time) has been satisfied after a previous PRECHARGE command to the identified bank; and
[0040] - After a previous ACTIVATE command, tRRD_L has been met (eg, tRRD_L must be met following ACTIVATE since REFpb will target banks in all bank groups).
[0041] It will be appreciated that during each bank refresh cycle time (tRFCpb), the target bank or the bank identified in the command is inaccessible. Therefore, the REFpb command will prevent access to the identified bank of tRFCpb when the bank performs a refresh operation. However, in one embodiment, other banks within the memory 120 are accessible and can be addressed during the refresh cycle (tRFCpb of the target bank). When each bank refresh cycle is completed, the affected bank will be in an idle state.
[0042] In one embodiment, after issuing the REFpb command, the following conditions must be met:
[0043] - Before issuing a REF command, tRFCpb must be satisfied;
[0044] - tRFCpb must be satisfied before issuing an ACTIVATE (ACT) command to the same bank; and
[0045] - Before issuing an ACT command or a REFpb command to a different bank, tRRD_L must be satisfied.
[0046] When the memory 120 applies refresh across all banks 134 in the bank grouping 132, it will be appreciated that the memory controller 110 is able to schedule and issue fewer refresh commands. For example, in a system with 8 banks, a single REF command may be replaced with 8 REFpb commands. Replacing a single REF command with multiple REFpb commands can provide increased control over scheduling, both selectively postponing and selectively increasing refresh command activity. In one embodiment, REFpb commands can be issued in bursts. The REFpb burst can be set with a limit determined using 8*{tRFCpb+(n-1)*tRRD_L}, where n is equal to the number of banks in the bank grouping, tRFCpb is the refresh cycle time per bank, and tRRD_L is a long row-to-row delay or delay time between activation commands to the same bank grouping.
[0047] Therefore, in a memory bank grouping architecture, each memory device includes multiple memory banks. In the memory bank grouping architecture, a memory controller can address a specific memory bank using a memory bank grouping identifier and a memory bank identifier, wherein the memory bank grouping identifier specifies the memory bank grouping, and the memory bank address specifies the memory bank within the memory bank grouping. Each memory bank grouping has multiple memory banks that are similarly addressed. For example, BG0 can have memory banks B0-B3, and BG1 can have memory banks B0-B3, which can be separately addressed by a combination of memory bank grouping and memory bank address information. The memory device and the memory controller can be coupled via I / O hardware. In one embodiment, the memory controller is configured to determine to send a command for execution on a portion of the memory, and specifically, for the identified memory bank. The memory controller sends or issues a command, and the memory device receives and decodes it. The command can be, for example, a precharge command or a refresh command, and identifies the memory bank. Depending on the implementation, identifying the memory bank grouping can be optional or undefined. In response to the command, the memory device can be configured to execute the command on all memory banks having the memory bank identifier (including memory banks in different memory bank groups).
[0048] Figure 22 is a representation of an embodiment of a command table for each bank refresh and all bank refresh commands and each bank precharge and all bank precharge commands. Command table 210 provides an example of an embodiment of commands used in a DDR4E or DDR5 system. The memory controller generates the listed commands to generate the desired operation. The memory controller sends the command to one or more connected memory devices (either directly connected or connected in a memory module) to trigger the memory device(s) to execute the command.
[0049] It will be appreciated that Table 210 only shows refresh commands and precharge commands, but typically all memory access commands are defined in a similar manner, shown as L for a logic low output (e.g., "0"), H for a logic high output (e.g., "1"), and V for don't care. Table 210 also shows values "BG" and "BA" that refer to a bank group address and a bank address, respectively. It will be appreciated that the number of bits used to specify an address can be different depending on how many banks and bank groups are supported in a particular memory system implementation. As shown, a system implementation command according to Table 210 supports up to eight bank groups, and up to four banks per bank group. In another implementation, for a system including a device having 4 bank groups and 8 banks, Table 210 can have BG0-BG1 and BA0-BA2. Other configurations are possible.
[0050] Command 212 shows a refresh all banks command with bit A12 set high. Because all banks are refreshed, the bank grouping bits and bank address bits are "don't care". In response to such a command, the memory device will refresh all banks. Command 214 shows a refresh each bank with bit A12 set low. In response to such a command, the memory device will refresh the bank at the indicated bank grouping (BG) and bank address (BA). In one embodiment, the bank group address is "don't care", and the memory device will refresh the bank identified by BA[1:0] in all bank groups.
[0051] Command 220 shows a precharge all (PREA) command, where bit A10 is set high. Because all banks are precharged, the bank grouping bit and bank address bit are "don't care". In response to such a command, the memory device will precharge all banks. Command 216 shows a precharge each bank (PREpb) command, where bit A10 is set low and bit A12 is set low. In response to such a command, the memory device will precharge the bank at the indicated bank address (BA) across all banks. In one embodiment, the bank group address is don't care, and the memory device will precharge the bank identified by BA[1:0] in all bank groups. Command 218 shows a precharge single bank (PRE) command, where bit A10 is set low and bit A12 is set high. In response to such a command, the memory device will precharge a specific bank identified by BG[2:0] and BA[1:0].
[0052] The implementation of each memory bank command can be different based on different system definitions. In one embodiment, the memory controller provides a command with such a memory bank identifier: the memory bank identifier sets a specific bit (e.g., an unused bit or one of the "don't care" bits) to indicate that each memory bank refresh command is to be applied to the identified memory bank in all memory bank groups. For example, consider that bit A10 can be used to identify whether each memory bank refresh is to be applied to all memory bank groups or only to a specified memory bank group. As another example, consider that bit A12 can be used to identify whether each memory bank precharge is to be applied to all memory bank groups or only to a specified memory bank group. Other implementations are possible. In one embodiment, the memory device is configured to apply REFpb and / or apply PREPb across all memory bank groups. For example, consider that bits A10 and A12 can be applied to refresh and / or precharge a specified memory bank, or refresh and / or precharge across all memory bank groups. Therefore, the memory controller can be configured not to drive the memory bank group identifier, and efficiently send commands without a memory bank group identifier. In one embodiment, in addition to being able to identify a specific bank grouping, the bank grouping identifier bits can be defined with a predefined code to allow all bank groups to be represented via the bank grouping bits. Alternatively, one of the other unused bits can be used as an all bank grouping identifier.
[0053] The bank grouping architecture requires a larger tCCD (CAS to CAS delay) or sequential CAS (column address strobe) command (or CAS to CAS) time, whether reading or writing, if the commands target the same bank grouping. In one embodiment, the tCCD across bank groups can be set equal to the burst length (BL), so that sequential data can be seamlessly streamed across bank groups. In one embodiment, the memory controller maps sequential addresses across bank groups. Therefore, when making access commands to sequential addresses, accesses occur across bank groups and allow the use of tCCD_S that is lower than tCCD_L, which is a longer delay between accesses to banks in the same bank grouping.
[0054] From the perspective of the memory controller, access across memory bank groups occupies different memory bank resources. Therefore, from the perspective of the memory controller, refreshing the same memory bank across memory bank groups will essentially occupy a single resource within the time tRFCpb. This refresh across different memory bank groups can reduce the command bandwidth problem on servers with multiple DIMMs (dual in-line memory modules) and memory blocks (ranks). By issuing refresh commands from the memory controller that the memory device executes on the memory bank across different memory bank groups, the memory controller can issue fewer refresh commands under certain conditions. For example, in one embodiment, the memory controller can issue 8 refresh commands when it should have issued 32 refresh commands instead, which refers to issuing 8 per-memory bank refreshes that are applied across 4 memory bank groups, rather than separately issuing 32 per-memory bank refreshes that refresh each of the 8 memory banks in each memory bank group.
[0055] Figure 3 300 shows the performance improvement demonstrated by testing the operation of refreshing the same memory bank across all memory bank groups in response to each memory bank refresh command. In the diagram 300, the relative performance shows the percentage of the performance difference with respect to the baseline. The labels on the horizontal axis of the diagram 300 indicate the application on which the test or performance is measured. It will be understood that the each memory bank refresh performance shown in the diagram 300 includes using the command to precharge and then refresh the same memory bank number across the memory bank grouping.
[0056] In the example of diagram 300, the baseline is 1x memory block refresh. 1x refers to a 7.8us tREF refresh rate for a 4Gb device using a tRFC of 260ns. Memory block refresh refers to performing refresh at the memory block level. The bar that goes into the negative percentage is marked as memory block 2x, and represents a performance loss with a 2x refresh rate of 3.9us. Therefore, the bar below the baseline assumes a standard memory block refresh, but the refresh rate is doubled. Each test has two bars extending into the positive percentage. The two bars represent the performance gain due to each memory bank refresh, where the refresh is performed across memory bank groups. The bar on the left represents each memory bank refresh with 7.8us tREFI. The bar on the right represents the performance gain using each memory bank refresh with 3.9ustREFI. Diagram 300 shows the loss of memory block refresh performance due to doubling the refresh rate, and even if the refresh rate is doubled, the loss of performance is more than compensated by using each memory bank refresh.
[0057] Figure 4 410. In one embodiment, a memory controller determines to refresh one or more memory banks in an associated memory device. In one embodiment, multiple memory devices (e.g., DRAM or SDRAM) are coupled to the same memory controller, and each memory device is capable of implementing the same command in parallel in separate memory bank groups and memory bank resources for the memory device. In one embodiment, the memory controller is configured with information indicating that the memory has a memory bank grouping architecture. In this implementation, in one embodiment, the memory controller can determine to send a command for execution only on an identified portion of the memory device. For example, the memory controller can determine to access only a specific memory bank of the memory, rather than accessing all memory banks. This operation can be supported by all memory bank commands and each memory bank command mechanisms according to any embodiment described herein.
[0058] In one embodiment, the memory controller generates a precharge command identifying a memory bank or banks to be precharged, 412. In one embodiment, the precharge command applies to the identified memory bank(s) across all memory bank groupings in the memory. The memory controller can send the precharge command to the memory, 414. In one embodiment, the memory receives the precharge command and decodes it, 416. In one embodiment, the memory precharges the identified memory bank(s) across all memory bank groupings in response to the precharge command, 418. The command can be sent according to any of the embodiments described herein, which can identify a precharge command that applies to all memory bank groupings rather than a single memory bank grouping.
[0059] In one embodiment, the memory controller generates a refresh command identifying the (multiple) memory banks to be refreshed, 420. Similar to what was stated above with respect to the precharge command, the command can be generated and sent according to any of the embodiments described herein. Thus, the memory controller can employ a mechanism to indicate whether the command is to be applied across all memory bank groupings or across a single memory bank grouping. Alternatively, the memory device can be configured to respond to all per-memory bank commands that apply to the identified memory banks in all memory bank groupings but not in a specific memory bank grouping. The memory controller can send the refresh command to the memory, 422. In one embodiment, the memory receives the refresh command and decodes it, 424. The memory can refresh the identified (multiple) memory banks across all memory bank groupings in response to the refresh command, 426.
[0060] In one embodiment, the memory controller increments a bank counter for the identified bank for all bank groupings, 428. In one embodiment, the memory controller maintains different bank counters for different bank groupings and is able to update all bank counters when issuing refresh commands that apply across bank groupings. In one embodiment, where the memory subsystem applies refreshes to the same bank across bank groupings, the memory controller maintains only a single per-bank refresh counter and updates it in conjunction with issuing each bank refresh that is applied across bank groupings. It will be appreciated that applying refresh commands that cover the same bank number across all bank groupings can provide good performance because accesses are interleaved across bank groupings to the same corresponding bank.
[0061] Figure 5 5 is a block diagram of an embodiment of a computing system in which a refresh of a specified memory bank can be implemented across memory bank groups. System 500 represents a computing device according to any embodiment described herein, and can be a laptop computer, a desktop computer, a server, a game or entertainment control system, a scanner, a copier, a printer, a routing or switching device, or other electronic device. System 500 includes a processor 520, which provides processing, operation management, and instruction execution of system 500. Processor 520 can include any type of microprocessor, central processing unit (CPU), processing core, or other processing hardware for providing processing of system 500. Processor 520 controls the overall operation of system 500, and can be or include one or more programmable general or special purpose microprocessors, digital signal processors (DSPs), programmable controllers, application specific integrated circuits (ASICs), programmable logic devices (PLDs), etc., or a combination of these devices.
[0062] The memory subsystem 530 represents the main memory of the system 500 and provides temporary storage for code to be executed by the processor 520 or data values to be used in executing routines. The memory subsystem 530 can include one or more memory devices, such as read-only memory (ROM), flash memory, one or more random access memories (RAM), or other memory devices, or a combination of these devices. The memory subsystem 530 stores and hosts (among other things) an operating system (OS) 536 to provide a software platform for executing instructions in the system 500. In addition, other instructions 538 from the memory subsystem 530 are stored and executed to provide the logic and processing of the system 500. The OS 536 and instructions 538 are executed by the processor 520. The memory subsystem 530 includes a memory device 532 that stores data, instructions, programs, or other items. In one embodiment, the memory subsystem includes a memory controller 534, which is a memory controller for generating commands and issuing commands to the memory device 532. It is understood that the memory controller 534 can be a physical part of the processor 520.
[0063] Processor 520 and memory subsystem 530 are coupled to bus / bus system 510. Bus 510 is an abstraction representing any one or more separate physical buses, communication lines / interfaces, and / or point-to-point connections connected by appropriate bridges, adapters, and / or controllers. Therefore, bus 510 can include, for example, one or more of the following: a system bus, a peripheral component interconnect (PCI) bus, a hypertransport or industry standard architecture (ISA) bus, a small computer system interface (SCSI) bus, a universal serial bus (USB), or an Institute of Electrical and Electronics Engineers (IEEE) standard 1394 bus (commonly referred to as "Firewire"). The bus of bus 510 can also correspond to an interface in network interface 550.
[0064] The system 500 also includes one or more input / output (I / O) interfaces 540, a network interface 550, one or more internal mass storage devices 560, and a peripheral interface 570 coupled to the bus 510. The I / O interface 540 can include one or more interface components (e.g., video, audio, and / or alphanumeric interfaces) through which a user interacts with the system 500. The network interface 550 provides the system 500 with the ability to communicate with remote devices (e.g., servers, other computing devices) over one or more networks. The network interface 550 can include an Ethernet adapter, a wireless interconnect component, a USB (Universal Serial Bus), or other wired or wireless standard-based or proprietary interfaces.
[0065] Storage device 560 can be or include any conventional media for storing large amounts of data in a non-volatile manner, such as one or more magnetic, solid-state, or optical based disks, or a combination thereof. Storage device 560 saves code or instructions and data 562 in a persistent state (i.e., although power to system 500 is interrupted, the value is retained). Storage device 560 can be generally considered as "memory", although memory 530 is an execution or operating memory for providing instructions to processor 520. Although storage device 560 is non-volatile, memory 530 can include volatile memory (i.e., if power to system 500 is interrupted, the value or state of the data is indeterminate).
[0066] Peripheral interface 570 can include any hardware interface not specifically mentioned above. Peripheral devices generally refer to devices that are dependently connected to system 500. A dependent connection is a connection where system 500 provides a software and / or hardware platform on which operations are performed or with which a user interacts.
[0067] In one embodiment, the memory subsystem 530 includes command logic 580, which can represent the logic of the memory 532 and / or the logic of the memory controller 534. The command logic 580 allows the memory subsystem to apply each memory bank command to the same identified memory bank across memory bank groups. For example, each memory bank command can include a per-memory bank precharge command and / or a per-memory bank refresh command. In one embodiment, each memory bank refresh command can identify a memory bank number and can refresh the same memory bank across all memory bank groups by the memory device. Similarly, in one embodiment, each memory bank precharge command can identify a memory bank number and can precharge the same memory bank across all memory bank groups by the memory device.
[0068] Figure 6 6 is a block diagram of an embodiment of a mobile device in which refreshing of a specified memory bank across memory bank groups can be implemented. Device 600 represents a mobile computing device, such as a computing tablet, a mobile phone or smart phone, a wireless-enabled e-reader, a wearable computing device, or other mobile device. It is understood that some of the components of such a device are generally shown, and not all components are shown in device 600.
[0069] Device 600 includes a processor 610 that performs the primary processing operations of device 600. Processor 610 can include one or more physical devices, such as a microprocessor, an application processor, a microcontroller, a programmable logic device, or other processing modules. The processing operations performed by processor 610 include the execution of an operating platform or operating system on which applications and / or device functions are executed. Processing operations include operations related to I / O (input / output) with a human user or with other devices, operations related to power management, and / or operations related to connecting device 600 to another device. Processing operations can also include operations related to audio I / O and / or display I / O.
[0070] In one embodiment, device 600 includes audio subsystem 620, which represents hardware (e.g., audio hardware and audio circuits) and software (e.g., drivers, codecs) components associated with providing audio functions to a computing device. Audio functions can include speaker and / or headphone outputs, and microphone inputs. Devices for such functions can be integrated into device 600, or connected to device 600. In one embodiment, a user interacts with device 600 by providing audio commands received and processed by processor 610.
[0071] The display subsystem 630 represents hardware (e.g., display device) and software (e.g., driver) components that provide visual and / or tactile display for user to interact with computing device. The display subsystem 630 includes a display interface 632, which includes a specific screen or hardware device for providing display to the user. In one embodiment, the display interface 632 includes logic separated from the processor 610 for performing at least some processing related to the display. In one embodiment, the display subsystem 630 includes a touch screen device that provides both output and input to the user. In one embodiment, the display subsystem 630 includes a high-definition (HD) display that provides output to the user. High definition can indicate that the display has a pixel density close to 100PPI (pixels per inch) or greater, and can include formats such as full HD (e.g., 1080p), retina display, 4K (ultra-high definition or UHD) or other.
[0072] I / O controller 640 represents the hardware devices and software components related to the interaction with the user. I / O controller 640 can operate to manage the hardware as a part of audio subsystem 620 and / or display subsystem 630. In addition, I / O controller 640 shows the connection point where additional devices are connected to device 600, and the user can interact with the system through this connection point. For example, the device that can be attached to device 600 can include a microphone device, a speaker or stereo system, a video system or other display device, a keyboard or keypad device, or other I / O devices used together with a specific application, such as a card reader or other device.
[0073] As mentioned above, the I / O controller 640 can interact with the audio subsystem 620 and / or the display subsystem 630. For example, input through a microphone or other audio device can provide input or commands for one or more applications or functions of the device 600. In addition, instead of or in addition to the display output, an audio output can be provided. In another example, if the display subsystem includes a touch screen, the display device also serves as an input device, which can be at least partially managed by the I / O controller 640. Additional buttons or switches can also be present on the device 600 to provide I / O functions managed by the I / O controller 640.
[0074] In one embodiment, I / O controller 640 manages devices such as accelerometers, cameras, light sensors or other environmental sensors, gyroscopes, global positioning systems (GPS), or other hardware that can be included in device 600. Input can be part of direct user interaction, as well as providing environmental input to the system to affect its operation (e.g., filtering noise, adjusting the display for brightness detection, applying a flash to a camera, or other features). In one embodiment, device 600 includes power management 650, which manages battery power utilization, battery charging, and features related to power saving operation.
[0075] The memory subsystem 660 includes (multiple) memory devices 662 for storing information in the device 600. The memory subsystem 660 may include non-volatile (the state does not change if power to the memory device is interrupted) memory devices and / or volatile (the state is indeterminate if power to the memory device is interrupted) memory devices. The memory 660 is capable of storing application data, user data, music, photos, documents, or other data related to the applications and functions of the execution system 600, as well as system data (whether long-term or temporary). In one embodiment, the memory subsystem 660 includes a memory controller 664 (which may also be considered part of the control of the system 600, and may potentially be considered part of the processor 610). The memory controller 664 includes a scheduler for generating commands and issuing commands to the memory devices 662.
[0076] Connections 670 include hardware devices (e.g., wireless and / or wired connectors and communication hardware) and software components (e.g., drivers, protocol stacks) that enable device 600 to communicate with external devices. External devices can be separate devices, such as other computing devices, wireless access points or base stations, and peripherals such as headsets, printers, or other devices.
[0077] Connection 670 can include a variety of different types of connections. In general, device 600 is shown as having a cellular connection 672 and a wireless connection 674. Cellular connection 672 generally refers to a cellular network connection provided by a wireless carrier, for example, via GSM (Global System for Mobile Communications) or variants or derivatives, CDMA (Code Division Multiple Access) or variants or derivatives, TDM (Time Division Multiplexing) or variants or derivatives, LTE (Long Term Evolution - also known as "4G"), or other cellular service standards. Wireless connection 674 refers to a non-cellular wireless connection, and can include a personal area network (e.g., Bluetooth), a local area network (e.g., WiFi), and / or a wide area network (e.g., WiMax), or other wireless communications. Wireless communication refers to transferring data through a non-solid medium using modulated electromagnetic radiation. Wired communication occurs through a solid communication medium.
[0078] Peripheral connections 680 include hardware interfaces and connectors, as well as software components (e.g., drivers, protocol stacks) that make peripheral connections. It will be appreciated that device 600 can be a peripheral device ("to" 682) to other computing devices, as well as have peripheral devices ("from" 684) connected thereto. Device 600 commonly has a "docking" connector that is used to connect to other computing devices for purposes such as managing (e.g., downloading and / or uploading, changing, synchronizing) content on device 600. Additionally, a docking connector can allow device 600 to connect to certain peripherals that allow device 600 to control output of content to, for example, an audio-visual or other system.
[0079] In addition to proprietary docking connectors or other proprietary connection hardware, the device 600 can make peripheral connections 680 via common or standards-based connectors. Common types can include Universal Serial Bus (USB) connectors (which can include any of a number of different hardware interfaces), DisplayPort including MiniDisplayPort (MDP), High-Definition Multimedia Interface (HDMI), FireWire, or other types.
[0080] In one embodiment, the memory subsystem 660 includes command logic 666, which can represent the logic of the memory 662 and / or the logic of the memory controller 664. The command logic 666 allows the memory subsystem to apply each memory bank command to the same identified memory bank across memory bank groups. For example, each memory bank command can include a per-memory bank precharge command and / or a per-memory bank refresh command. In one embodiment, each memory bank refresh command can identify the memory bank number and can refresh the same memory bank across all memory bank groups by the memory device. Similarly, in one embodiment, each memory bank precharge command can identify the memory bank number and can precharge the same memory bank across all memory bank groups by the memory device.
[0081] In one aspect, a memory device for refreshing in a memory subsystem comprises: multiple memory banks of a memory in a memory bank grouping architecture, wherein the memory banks of a memory resource are separately addressable in a memory bank grouping by a memory bank grouping identifier and a memory bank identifier; I / O (input / output) hardware coupled to a memory controller and receiving a command from the memory controller, the command including a memory bank identifier; and logic for decoding the command and accessing a memory bank corresponding to the memory bank identifier based at least in part on the command, the memory bank including memory banks in different memory bank groups.
[0082] In one embodiment, the I / O hardware is configured to receive commands with memory bank identifiers for more than one memory bank. In one embodiment, the I / O hardware is configured to receive commands without memory bank group identifiers. In one embodiment, the I / O hardware is configured to receive commands with memory bank group identifiers for selecting all memory bank groups. In one embodiment, the commands include refresh commands. In one embodiment, the commands include precharge commands. In one embodiment, the commands include a sequence of precharge commands and refresh commands for accessing the same memory bank identifier in different memory bank groups. In one embodiment, a memory controller coupled to the memory device is also included. In one embodiment, at least one processor and a network interface are also included, the at least one processor is communicatively coupled to the memory controller, and the network interface is communicatively coupled to the at least one processor.
[0083] In one aspect, a system for performing a refresh includes: a memory controller; and a memory device coupled to the memory controller, the memory device including: multiple memory banks of a memory in a memory bank grouping architecture, in which the memory banks of a memory resource are separately addressable in a memory bank grouping by a memory bank grouping identifier and a memory bank identifier; I / O (input / output) hardware coupled to the memory controller and receiving a command from the memory controller, the command including the memory bank identifier; and logic for decoding the command and accessing a memory bank corresponding to the memory bank identifier based at least in part on the command, the memory bank including memory banks in different memory bank groups.
[0084] In one aspect of the system, the memory device is according to any embodiment of the above memory device aspect. In one embodiment, it also includes at least one processor communicatively coupled to the memory controller. In one embodiment, the memory controller is part of the processor. In one embodiment, it also includes a network interface communicatively coupled to the at least one processor.
[0085] In one aspect, a method for refresh in a memory subsystem includes: accepting a command from a memory controller at a memory device, the command including a memory body identifier, the memory device having a memory body grouping architecture, in which memory bodies of a memory resource can be separately addressed in a memory body group by a memory body group identifier and a memory body identifier; and in response to receiving the command, causing all memory bodies corresponding to the memory body identifier to execute the command, the memory bodies including memory bodies in different memory body groups.
[0086] In one aspect of the method, the method includes operations according to any of the embodiments set forth above with respect to a memory device. In one aspect, an article of manufacture includes a computer-readable storage medium having content stored thereon, the content, when accessed, causing execution of operations to perform a method for refreshing according to any of the embodiments of the method. In one aspect, an apparatus for refreshing includes modules for executing operations to perform a method according to any of the embodiments of the method.
[0087] In one aspect, a memory controller includes: command logic that determines to send a command for execution on a portion of an associated memory device to access only an identified portion of the memory device, the memory device having a memory bank grouping architecture in which memory banks of a memory resource are separately addressable in a memory bank group by a memory bank group identifier and a memory bank identifier; and I / O (input / output) hardware that is coupled to the memory device, the I / O hardware generates a command and sends the command to the memory device, the command including the memory bank identifier, the command accessing all memory banks within the memory device corresponding to the memory bank identifier for executing the command, the memory banks including memory banks in different memory bank groups.
[0088] In one embodiment, the I / O hardware sending a command includes the I / O hardware being configured to send a command with a bank identifier for more than one bank. In one embodiment, the I / O hardware sending a command includes the I / O hardware being configured to send a command without a bank grouping identifier. In one embodiment, the I / O hardware sending a command includes the I / O hardware being configured to send a command with a bank grouping identifier for selecting all bank groups. In one embodiment, the I / O hardware sending a command includes the I / O hardware being configured to sequentially send a precharge command, a refresh command, or both for accessing the same bank identifier in different bank groups.
[0089] In one aspect, a second system for performing refresh includes: a memory device; and a memory controller coupled to the memory device, the memory controller including command logic, the command logic determining to send a command for execution on a portion of an associated memory device to access only an identified portion of the memory device, the memory device having a memory bank grouping architecture, in which memory banks of a memory resource are separately addressable in a memory bank group by a memory bank group identifier and a memory bank identifier; and I / O (input / output) hardware coupled to the memory device, the I / O hardware generating a command and sending the command to the memory device, the command including a memory bank identifier, the command accessing all memory banks within the memory device corresponding to the memory bank identifier for executing the command, the memory banks including memory banks in different memory bank groups.
[0090] In one aspect of the second system, the memory device is according to any embodiment of the above memory controller aspect. In one embodiment, it also includes at least one processor communicatively coupled to the memory controller. In one embodiment, the memory controller is part of the processor. In one embodiment, it also includes a network interface communicatively coupled to the at least one processor.
[0091] In one aspect, a second method for refresh in a memory subsystem includes: determining to send a command for execution on a portion of an associated memory device to access only an identified portion of the memory device, the memory device having a memory bank grouping architecture in which memory banks of a memory resource are separately addressable in a memory bank group by a memory bank grouping identifier and a memory bank identifier; and sending a command to the memory device, the command including the memory bank identifier, the command accessing all memory banks within the memory device corresponding to the memory bank identifier for executing the command, the memory banks including memory banks in different memory bank groups.
[0092] In one aspect of the second method, the second method includes operations according to any of the embodiments set forth above with respect to a memory controller. In one aspect, an article of manufacture includes a computer-readable storage medium having content stored thereon, the content, when accessed, causing operations to be performed to perform a method for refreshing according to any of the embodiments of the second method. In one aspect, an apparatus for refreshing includes a module for performing operations to perform a method according to any of the embodiments of the second method.
[0093] Flowcharts as shown herein provide examples of sequences of various processing actions. Flowcharts can indicate operations to be performed by software or firmware routines, as well as physical operations. In one embodiment, a flow chart can illustrate the state of a finite state machine (FSM) that can be implemented in hardware and / or software. Although shown in a specific sequence or order, the order of actions can be modified unless otherwise stated. Therefore, the embodiments shown should only be understood as examples, and the process can be performed in different orders, and some actions can be performed in parallel. In addition, one or more actions can be omitted in various embodiments; therefore, not all actions are required in each embodiment. Other process flows are possible.
[0094] Within the scope of the various operations or functions described herein, they can be described or defined as software code, instructions, configuration, and / or data. The content can be source code or differential code ("delta" or "patch" code) that is directly executable ("object" or "executable" form). The software content of the embodiments described herein can be provided via an article having content stored thereon, or via a method of operating a communication interface to send data via a communication interface. The machine-readable storage medium enables the machine to perform the described functions or operations, and includes any mechanism for storing information in a form accessible to a machine (e.g., a computing device, an electronic system, etc.), such as a recordable / unrecordable medium (e.g., a read-only memory (ROM), a random access memory (RAM), a disk storage medium, an optical storage medium, a flash memory device, etc.). The communication interface includes any mechanism for communicating to another device with any medium interface in a hardwired, wireless, optical, etc. medium, such as a memory bus interface, a processor bus interface, an Internet connection, a disk controller, etc. The communication interface can be configured by providing configuration parameters and / or sending signals to prepare the communication interface for providing data signals describing the software content. The communication interface can be accessed via one or more commands or signals sent to the communication interface.
[0095] The various components described herein can be modules for performing the described operations or functions. Each component described herein includes software, hardware, or a combination thereof. These components can be implemented as software modules, hardware modules, dedicated hardware (e.g., application-specific hardware, application-specific integrated circuits (ASICs), digital signal processors (DSPs), etc.), embedded controllers, hard-wired circuits, etc.
[0096] In addition to what is described herein, various modifications can be made to the disclosed embodiments and implementations of the present invention without departing from the scope of the present invention. Therefore, the descriptions and examples herein should be interpreted as illustrative rather than limiting. The scope of the present invention should be measured only by reference to the appended claims.
Claims
1. A memory device having a memory bank grouping architecture, comprising: a plurality of memory bank groups, each memory bank group comprising a plurality of memory banks, the memory banks being separately addressable by a combination of a memory bank group identifier and a memory bank identifier, wherein the memory bank group identifier is used to specify the memory bank group, and the memory bank identifier is used to specify the memory bank within the memory bank group; I / O hardware coupled to the memory controller and receiving a refresh command from the memory controller, wherein a specific address bit in the refresh command is used to indicate whether the refresh command is used to refresh all of the memory banks or to refresh designated memory banks in all of the memory bank groups in parallel; and logic for: decoding the refresh command, and when the refresh command is for refreshing all the memory banks, refreshing all the memory banks, and when the refresh command is for refreshing designated memory banks in all the memory bank groups, refreshing designated memory banks in all the memory bank groups in parallel, In which, in the case where the refresh command is used to refresh the specified storage body in all the storage body groups, the refresh command includes the storage body identifier and does not have the storage body group identifier, and the refresh command is applied to the storage bodies with the storage body identifier in all the storage body groups.
2. The memory device having a bank grouping architecture according to claim 1, wherein the memory device comprises a synchronous dynamic random access memory (SDRAM) compatible with a double data rate version 5 (DDR5) standard.
3. The memory device having a bank grouping architecture according to claim 1 , further comprising: A memory bank counter is used to be updated when each of the memory banks is refreshed.
4. A memory controller comprising: command logic, the command logic generating a refresh command for refreshing one or more memory banks in an associated memory device, the memory device comprising: a plurality of memory bank groups, each memory bank group comprising a plurality of the memory banks, the memory banks being separately addressable by a combination of a memory bank group identifier and a memory bank identifier, wherein the memory bank group identifier is used to specify the memory bank group, and the memory bank identifier is used to specify a memory bank within the memory bank group; I / O hardware coupled to the memory device, the I / O hardware sending the refresh command to the memory device, the specific address bit in the refresh command being used to indicate whether the refresh command is used to refresh all of the memory banks or to refresh designated memory banks in all of the memory bank groups in parallel, In which, in the case where the refresh command is used to refresh the specified storage body in all the storage body groups, the refresh command includes the storage body identifier and does not have the storage body group identifier, and the refresh command is applied to the storage bodies with the storage body identifier in all the storage body groups.
5. The memory controller of claim 4, the memory device comprising a synchronous dynamic random access memory (SDRAM) compliant with a double data rate version 5 (DDR5) standard.
6. A storage system comprising: a memory controller configured to manage refresh of an associated memory device; as well as A memory device having a bank grouping architecture is coupled to the memory controller. Wherein, the memory device comprises: a plurality of memory bank groups, each memory bank group comprising a plurality of memory banks, the memory banks being separately addressable by a combination of a memory bank group identifier and a memory bank identifier, wherein the memory bank group identifier is used to specify the memory bank group, and the memory bank identifier is used to specify the memory bank within the memory bank group; I / O hardware coupled to the memory controller and receiving a refresh command from the memory controller, wherein a specific address bit in the refresh command is used to indicate whether the refresh command is used to refresh all of the memory banks or to refresh a specified memory bank in all of the memory bank groups; and logic for: decoding the refresh command, and when the refresh command is for refreshing all the memory banks, refreshing all the memory banks, and when the refresh command is for refreshing designated memory banks in all the memory bank groups, refreshing designated memory banks in all the memory bank groups in parallel, In which, in the case where the refresh command is used to refresh the specified storage body in all the storage body groups, the refresh command includes the storage body identifier and does not have the storage body group identifier, and the refresh command is applied to the storage bodies with the storage body identifier in all the storage body groups.
7. The memory system of claim 6, wherein the memory device comprises a synchronous dynamic random access memory (SDRAM) compliant with a double data rate version 5 (DDR5) standard.
8. The storage system according to claim 6, wherein the storage device further comprises: A memory bank counter is used to be updated when each of the memory banks is refreshed.
