WO2002015019A1 - Method and system for using dynamic random access memory as cache memory - Google Patents

Method and system for using dynamic random access memory as cache memory Download PDF

Info

Publication number
WO2002015019A1
WO2002015019A1 PCT/US2001/025494 US0125494W WO0215019A1 WO 2002015019 A1 WO2002015019 A1 WO 2002015019A1 US 0125494 W US0125494 W US 0125494W WO 0215019 A1 WO0215019 A1 WO 0215019A1
Authority
WO
WIPO (PCT)
Prior art keywords
block
dram
data
sram
array
Prior art date
Application number
PCT/US2001/025494
Other languages
French (fr)
Inventor
Brent Keeth
Brian M. Shirley
Charles H. Dennison
Kevin J. Ryan
Original Assignee
Micron Technology, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Micron Technology, Inc. filed Critical Micron Technology, Inc.
Priority to AU2001284924A priority Critical patent/AU2001284924A1/en
Publication of WO2002015019A1 publication Critical patent/WO2002015019A1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0893—Caches characterised by their organisation or structure
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C11/00—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor
    • G11C11/21—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements
    • G11C11/34—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices
    • G11C11/40—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors
    • G11C11/401—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors forming cells needing refreshing or charge regeneration, i.e. dynamic cells
    • G11C11/406—Management or control of the refreshing or charge-regeneration cycles
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C11/00—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor
    • G11C11/21—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements
    • G11C11/34—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices
    • G11C11/40—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors
    • G11C11/401—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using electric elements using semiconductor devices using transistors forming cells needing refreshing or charge regeneration, i.e. dynamic cells
    • G11C11/4063—Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing or timing
    • G11C11/407—Auxiliary circuits, e.g. for addressing, decoding, driving, writing, sensing or timing for memory cells of the field-effect type
    • G11C11/409—Read-write [R-W] circuits 
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C15/00—Digital stores in which information comprising one or more characteristic parts is written into the store and in which information is read-out by searching for one or more of these characteristic parts, i.e. associative or content-addressed stores
    • G11C15/04—Digital stores in which information comprising one or more characteristic parts is written into the store and in which information is read-out by searching for one or more of these characteristic parts, i.e. associative or content-addressed stores using semiconductor elements
    • G11C15/043—Digital stores in which information comprising one or more characteristic parts is written into the store and in which information is read-out by searching for one or more of these characteristic parts, i.e. associative or content-addressed stores using semiconductor elements using capacitive charge storage elements
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
    • G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
    • G11C7/1006—Data managing, e.g. manipulating data before writing or reading out, data bus switches or control circuits therefor
    • G—PHYSICS
    • G11—INFORMATION STORAGE
    • G11C—STATIC STORES
    • G11C7/00—Arrangements for writing information into, or reading information out from, a digital store
    • G11C7/10—Input/output [I/O] data interface arrangements, e.g. I/O data control circuits, I/O data buffers
    • G11C7/1015—Read-write modes for single port memories, i.e. having either a random port or a serial port
    • G11C7/1042—Read-write modes for single port memories, i.e. having either a random port or a serial port using interleaving techniques, i.e. read-write of one part of the memory while preparing another part
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0844—Multiple simultaneous or quasi-simultaneous cache accessing
    • G06F12/0846—Cache with multiple tag or data arrays being simultaneously accessible
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0893—Caches characterised by their organisation or structure
    • G06F12/0897—Caches characterised by their organisation or structure with two or more cache hierarchy levels

Definitions

  • the present invention is directed memory devices, and, more particularly, to a system and method for allowing dynamic random access memory devices to be used as cache memory.
  • DRAM dynamic random access memory
  • the cache memory stores instructions and/or data (collectively referred to as "data") that is frequently accessed by the processor or similar device, and may be accessed substantially faster than data can be accessed in system memory. It is important for the processor or similar device to be able to access the cache memory as needed. If the cache memory cannot be accessed for a period, the operation of the processor or similar device must be halted during this period.
  • Cache memory is typically implemented using static random access memory (“SRAM”) because such memory need not be refreshed and is thus always accessible for a write or a read memory access.
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • a method of caching data and a cache system that may be used in a computer system includes a DRAM having a plurality of refresh blocks and a pair of SRAMs having a capacity of at least the capacity of the refresh blocks. If a block of the DRAM to which data is attempting to be written is being refreshed, the data is instead written to one of the SRAMs. When the refresh of that block has been completed, the data is transferred from the SRAM to a block of the DRAM to which data was attempted to be written. If a block to which data is attempting to be written is being refreshed and data is being transferred from the one SRAM to a block of the DRAM, the data is instead written to the other SRAM. As a result, there is always one SRAM available into which data may be written if a refresh block to which the write was directed is being refreshed. BRIEF DESCRIPTION OF THE DRAWINGS
  • FIG. 1 is a block diagram of a computer system containing a cache memory in accordance with one embodiment of the invention.
  • FIG. 2 is a block diagram of a cache system that may be used as a cache memory in the computer system of Figure 2 in accordance with one embodiment of the invention.
  • Figure 3 is a diagram conceptually illustrating a DRAM and SRAM arrays shown in the cache system of Figure 2.
  • FIG. 1 is a block diagram of a computer system 10 that includes a processor 12 for performing various computing functions by executing software to perform specific calculations or tasks.
  • the processor 12 is coupled to a processor bus 14 that normally includes an address bus, a control bus, and a data bus (not separately shown).
  • the computer system 10 includes a system memory 16, which is typically dynamic random access memory ("DRAM"). As mentioned above, using DRAM at the system memory 16 provides relatively high capacity at relatively little expense.
  • the system memory 16 is coupled to the processor bus 14 by a system controller 20 or similar device, which is also coupled to an expansion bus 22, such as a Peripheral Component Interface (“PCI”) bus.
  • PCI Peripheral Component Interface
  • a bus 26 coupling the system controller 20 to the system memory 16 also normally includes an address bus, a control bus, and a data bus (not separately shown), although other architectures can be used.
  • the data bus of the system memory 16 may be coupled to the data bus of the processor bus 14, or the system memory 16 may be implemented by a packetized memory (not shown), which normally does not include a separate address bus and control bus.
  • the computer system 10 also includes one or more input devices 34, such as a keyboard or a mouse, coupled to the processor 12 through the expansion bus 22, the system controller 20, and the processor bus 14. Also typically coupled to the expansion bus 22 are one or more output devices 36, such as a printer or a video terminal.
  • One or more data storage devices 38 are also typically coupled to the expansion bus 22 to allow the processor 12 to store data or retrieve data from internal or external storage media (not shown). Examples of typical storage devices 38 include hard and floppy disks, tape cassettes, and compact disk read-only memories (CD- ROMs).
  • the processor 12 is also typically coupled to cache memory 40 through the processor bus 14.
  • the cache memory 40 was normally implemented using static random access memory (“SRAM”) because such memory is relatively fast, and does not require refreshing and may thus always be accessed.
  • SRAM static random access memory
  • using SRAM for the cache memory 40 is a relatively expensive means for providing a relatively high capacity because of the large number of components making up each SRAM storage cell compared to the number of components in each DRAM storage cell.
  • the cache memory 40 shown in Figure 1 is implemented using a cache system 50, an example of which is shown in Figure 2.
  • the cache system 50 includes components normally found in a DRAM, including an address decoder 52 receiving addresses through an address bus 53, a row driver circuit 54 adapted to receive row addresses from the address decoder 52, and a column driver circuit 56 adapted to receive column addresses from the address decoder 52.
  • the row driver circuit 54 is coupled to word lines (not shown) in a memory array 60
  • the column driver circuit 56 is coupled to digit lines (not shown) in the memory array 60.
  • the memory array 60 is either physically or logically divided into a plurality of banks 60a-n.
  • Each bank 60a-n is divided into one or more refresh blocks, each containing a plurality of rows that are contemporaneously refreshed.
  • the column driver 56 is also coupled to a sense amplifier/write driver circuit 64 to route write data and read data from and to, respectively, a data input/output buffer 66 through an internal data bus 68.
  • the data input/output buffer 66 is, in turn, coupled to an external data bus 70.
  • the cache system 50 also includes a control circuit 72 that includes a command buffer 74 receiving command signals through a command bus 76 and generating appropriate control signals for controlling the operation of the cache system 50.
  • the control circuit 72 also includes a refresh controller 78 for refreshing the DRAM array 60 one refresh block at a time.
  • the cache system 50 also includes two SRAM arrays 80, 84 that are each coupled to the sense amplifier/write driver circuit 64 to access data in the DRAM array 60.
  • the SRAM arrays 80, 84 are also coupled to the refresh controller 78.
  • the refresh controller 78 receives addresses from the address decoder 52, and it applies addressing and control signals to the row driver 54.
  • each of the refresh blocks 61a-n has a capacity of Y bits
  • each of the SRAM arrays 80, 84 also has a capacity of Y bits.
  • Each of the refresh blocks 61a-n may be individually refreshed under control of the refresh controller 78 ( Figure 2).
  • the DRAM array 60 has twice the normal number of input/output ("I/O") lines, which are configured so that two blocks can be simultaneously accessed. As a result, it is possible for data to be read from or written to one refresh block 61a-n of the DRAM array 60 at the same time data are being transferred from one of the SRAM arrays 80, 84 to another block 61a-n of the DRAM array 60.
  • I/O input/output
  • a read from a refresh block 61a-n that is not being refreshed is read in a conventional manner.
  • a write to a block 61a-n that is not being refreshed is accomplished in a conventional manner.
  • no problem is presented in either writing to or reading from a refresh block 61a-n that is not being refreshed.
  • data access to the cache system 50 does not require any wait, thus allowing the cache system 50 to be used as a cache memory in place of a typically used SRAM without any performance limitations.
  • the potential problem in accessing the cache system 50 is in the event of a read or a write to a refresh block 61a-n being refreshed, and, in particular, to a different row than the row in that block that is being refreshed.
  • the cache system 50 preferably the refresh controller 78, may check each memory command prior to initiating a refresh in a block 61a-n to determine if the memory command is a read. If a read command directed to a block 61a-n that is about to be refreshed is received, then the refresh is not initiated. In this regard, it is assumed that the duration of a refresh is shorter than the duration of a memory read operation.
  • the read data are written to one of the SRAMs 82, 84.
  • the read data are subsequently accessible in one of the SRAMs 82, 84, thereby allowing the portion of the block 61 a-n that stored such data to be refreshed despite subsequent reads from that that portion.
  • the reads will refresh the rows.
  • the write data is instead written to one of the SRAM arrays 80, 84.
  • the refresh controller 78 starts a refresh of another block 61a-n of the DRAM array 60. While this subsequent refresh is occurring, the data that had been written to one of the SRAM arrays 80, 84 is transferred to the block 61 a-n to which the earlier write was directed. If, during refresh of the second block 61 a-n, a read or a write is directed toward that block 61 a-n, then that data is instead stored in the other one of the SRAM arrays 80, 84.
  • the cache system 50 may therefore be used as a cache memory in place of an SRAM that is typically used, thereby providing high capacity caching at relatively little cost.

Abstract

A cache memory system includes a DRAM (60) having a plurality of banks and 2 SRAMs (80, 84). In operation, data read from a bank (60a) of the DRAM are stored in one of the SRAMs so that repeated hits to the bank are cached by reading from the SRAM. In the event of write to a bank being refreshed, the write data are stored in one of the SRAMs. After the refresh of the bank has been completed, the data stored in the SRAM are transfered to the bank. A subsequent read or write to a second bank undergoing refresh and occurring during the transfer of data from an SRAM (80) to the DRAM is stored in the second bank. If the second bank is being refreshed, the data are stored in the other SRAM (84).

Description

METHOD AND SYSTEM FOR USING DYNAMIC RANDOM ACCESS MEMORY
AS CACHE MEMORY
TECHNICAL FIELD
The present invention is directed memory devices, and, more particularly, to a system and method for allowing dynamic random access memory devices to be used as cache memory.
BACKGROUND OF THE INVENTION
Memory devices are used in a wide variety of applications, including computer systems. Computer systems and other electronic devices containing a microprocessor or similar device typically include system memory, which is generally implemented using dynamic random access memory ("DRAM"). The primary advantage of DRAM is that it uses relatively few components to store each bit of data, and is thus relatively inexpensive to provide relatively high capacity system memory. A disadvantage of DRAM, however, is that their memory cells must be periodically refreshed. While a memory cell is being refreshed, read and write accesses to other rows in the memory array are blocked. The need to refresh memory cells does not present a significant problem in most applications, but it can prevent their use in applications where immediate access to memory cells is required or highly desirable.
Also included in many computer systems and other electronic devices is a cache memory. The cache memory stores instructions and/or data (collectively referred to as "data") that is frequently accessed by the processor or similar device, and may be accessed substantially faster than data can be accessed in system memory. It is important for the processor or similar device to be able to access the cache memory as needed. If the cache memory cannot be accessed for a period, the operation of the processor or similar device must be halted during this period. Cache memory is typically implemented using static random access memory ("SRAM") because such memory need not be refreshed and is thus always accessible for a write or a read memory access. However, a significant disadvantage of SRAM is that each memory cell requires a relatively large number of components, thus making SRAM data storage relatively expensive. It would be desirable to implement cache memory using DRAM because high capacity cache memories could then be provided at relatively little cost. However, a cache memory implemented using DRAMs would be inaccessible at certain times during a refresh of the memory cells in the DRAM. For example, during refresh of a row of memory cells, it would be impossible to read data from or write data to other rows of memory cells. As a result of these problems, DRAMs have not generally been considered acceptable for use as cache memory or for other applications requiring immediate access to memory.
Attempts have been made to use DRAM as cache memory, but these attempts have not been entirely successful in solving the refresh problem so that these prior art devices are not always available for a memory access. These prior art devices have attempted to "hide" memory refreshes by including a small SRAM to store one or more rows of DRAM data during refresh of a row being addressed. However, in practice, there are still some memory access situations in which these prior art devices may not be accessed, thus suspending the operation of a processor or similar device. There is therefore a need for a DRAM that effectively hides memory refresh under all memory access situations so that the DRAM may provide relatively inexpensive, high capacity cache memory.
SUMMARY OF THE INVENTION
A method of caching data and a cache system that may be used in a computer system includes a DRAM having a plurality of refresh blocks and a pair of SRAMs having a capacity of at least the capacity of the refresh blocks. If a block of the DRAM to which data is attempting to be written is being refreshed, the data is instead written to one of the SRAMs. When the refresh of that block has been completed, the data is transferred from the SRAM to a block of the DRAM to which data was attempted to be written. If a block to which data is attempting to be written is being refreshed and data is being transferred from the one SRAM to a block of the DRAM, the data is instead written to the other SRAM. As a result, there is always one SRAM available into which data may be written if a refresh block to which the write was directed is being refreshed. BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a block diagram of a computer system containing a cache memory in accordance with one embodiment of the invention.
Figure 2 is a block diagram of a cache system that may be used as a cache memory in the computer system of Figure 2 in accordance with one embodiment of the invention.
Figure 3, is a diagram conceptually illustrating a DRAM and SRAM arrays shown in the cache system of Figure 2.
DETAILED DESCRIPTION OF THE INVENTION ' Figure 1 is a block diagram of a computer system 10 that includes a processor 12 for performing various computing functions by executing software to perform specific calculations or tasks. The processor 12 is coupled to a processor bus 14 that normally includes an address bus, a control bus, and a data bus (not separately shown). In addition, the computer system 10 includes a system memory 16, which is typically dynamic random access memory ("DRAM"). As mentioned above, using DRAM at the system memory 16 provides relatively high capacity at relatively little expense. The system memory 16 is coupled to the processor bus 14 by a system controller 20 or similar device, which is also coupled to an expansion bus 22, such as a Peripheral Component Interface ("PCI") bus. A bus 26 coupling the system controller 20 to the system memory 16 also normally includes an address bus, a control bus, and a data bus (not separately shown), although other architectures can be used. For example, the data bus of the system memory 16 may be coupled to the data bus of the processor bus 14, or the system memory 16 may be implemented by a packetized memory (not shown), which normally does not include a separate address bus and control bus. The computer system 10 also includes one or more input devices 34, such as a keyboard or a mouse, coupled to the processor 12 through the expansion bus 22, the system controller 20, and the processor bus 14. Also typically coupled to the expansion bus 22 are one or more output devices 36, such as a printer or a video terminal. One or more data storage devices 38 are also typically coupled to the expansion bus 22 to allow the processor 12 to store data or retrieve data from internal or external storage media (not shown). Examples of typical storage devices 38 include hard and floppy disks, tape cassettes, and compact disk read-only memories (CD- ROMs).
The processor 12 is also typically coupled to cache memory 40 through the processor bus 14. In the past, the cache memory 40 was normally implemented using static random access memory ("SRAM") because such memory is relatively fast, and does not require refreshing and may thus always be accessed. However, as explained above, using SRAM for the cache memory 40 is a relatively expensive means for providing a relatively high capacity because of the large number of components making up each SRAM storage cell compared to the number of components in each DRAM storage cell.
According to one embodiment of the invention, the cache memory 40 shown in Figure 1 is implemented using a cache system 50, an example of which is shown in Figure 2. The cache system 50 includes components normally found in a DRAM, including an address decoder 52 receiving addresses through an address bus 53, a row driver circuit 54 adapted to receive row addresses from the address decoder 52, and a column driver circuit 56 adapted to receive column addresses from the address decoder 52. The row driver circuit 54 is coupled to word lines (not shown) in a memory array 60, and the column driver circuit 56 is coupled to digit lines (not shown) in the memory array 60. As shown in Figure 2, the memory array 60 is either physically or logically divided into a plurality of banks 60a-n. Each bank 60a-n is divided into one or more refresh blocks, each containing a plurality of rows that are contemporaneously refreshed. The column driver 56 is also coupled to a sense amplifier/write driver circuit 64 to route write data and read data from and to, respectively, a data input/output buffer 66 through an internal data bus 68. The data input/output buffer 66 is, in turn, coupled to an external data bus 70. As in conventional DRAMs, the cache system 50 also includes a control circuit 72 that includes a command buffer 74 receiving command signals through a command bus 76 and generating appropriate control signals for controlling the operation of the cache system 50. The control circuit 72 also includes a refresh controller 78 for refreshing the DRAM array 60 one refresh block at a time.
Unlike conventional DRAMs, the cache system 50 also includes two SRAM arrays 80, 84 that are each coupled to the sense amplifier/write driver circuit 64 to access data in the DRAM array 60. The SRAM arrays 80, 84 are also coupled to the refresh controller 78. The refresh controller 78 receives addresses from the address decoder 52, and it applies addressing and control signals to the row driver 54.
The operation of the command buffer 74, refresh controller 78 and the SRAM arrays 80, 84 in relation to the other components of the cache system 50 will now be explained with reference to the diagram of Figure 3, which conceptually illustrates the DRAM array 60 and the SRAM arrays 80, 84 shown in Figure 2. As mentioned above, the DRAM array is divided into a plurality of refresh blocks. The refresh blocks may be part of the same or different banks 60a-n of DRAM memory, or physically different DRAM devices. In the embodiment shown in Figure 3, each of the refresh blocks 61a-n has a capacity of Y bits, and each of the SRAM arrays 80, 84 also has a capacity of Y bits. Each of the refresh blocks 61a-n may be individually refreshed under control of the refresh controller 78 (Figure 2). The DRAM array 60 has twice the normal number of input/output ("I/O") lines, which are configured so that two blocks can be simultaneously accessed. As a result, it is possible for data to be read from or written to one refresh block 61a-n of the DRAM array 60 at the same time data are being transferred from one of the SRAM arrays 80, 84 to another block 61a-n of the DRAM array 60.
In operation, a read from a refresh block 61a-n that is not being refreshed is read in a conventional manner. Similarly, a write to a block 61a-n that is not being refreshed is accomplished in a conventional manner. Thus, no problem is presented in either writing to or reading from a refresh block 61a-n that is not being refreshed. In either of these cases, data access to the cache system 50 does not require any wait, thus allowing the cache system 50 to be used as a cache memory in place of a typically used SRAM without any performance limitations.
The potential problem in accessing the cache system 50 is in the event of a read or a write to a refresh block 61a-n being refreshed, and, in particular, to a different row than the row in that block that is being refreshed. The cache system 50, preferably the refresh controller 78, may check each memory command prior to initiating a refresh in a block 61a-n to determine if the memory command is a read. If a read command directed to a block 61a-n that is about to be refreshed is received, then the refresh is not initiated. In this regard, it is assumed that the duration of a refresh is shorter than the duration of a memory read operation. Each time a read is executed, the read data are written to one of the SRAMs 82, 84. As a result, the read data are subsequently accessible in one of the SRAMs 82, 84, thereby allowing the portion of the block 61 a-n that stored such data to be refreshed despite subsequent reads from that that portion. In the case of sequential reads from the rows of a block 61 a-n, the reads will refresh the rows.
In the event a memory access is a write to a block 61 a-n being refreshed, the write data is instead written to one of the SRAM arrays 80, 84. When the refresh of the block to which the write was directed has been completed, the refresh controller 78 starts a refresh of another block 61a-n of the DRAM array 60. While this subsequent refresh is occurring, the data that had been written to one of the SRAM arrays 80, 84 is transferred to the block 61 a-n to which the earlier write was directed. If, during refresh of the second block 61 a-n, a read or a write is directed toward that block 61 a-n, then that data is instead stored in the other one of the SRAM arrays 80, 84. By the time the refresh of the second block 61 a-n has been completed, transfer of the data from first one of the SRAM arrays 80, 84 to the first block 61 a-n will have been completed, and that SRAM array 80, 84 will be available to store write data that is subsequently directed to any other block 61 a-n that is being refreshed. Therefore, an SRAM array 80, 84 is always available to store write data that is directed to a refresh block 61 a-n of the memory array 60 that is being refreshed. As a result, data may always be read from or written to the cache system 50 without the need for to wait for the completion of a refresh of any block 61 a-n the cache system 50. The cache system 50 may therefore be used as a cache memory in place of an SRAM that is typically used, thereby providing high capacity caching at relatively little cost. From the foregoing it will be appreciated that, although specific embodiments of the invention have been described herein for purposes of illustration, various modifications may be made without deviating from the spirit and scope of the invention. Accordingly, the invention is not limited except as by the appended claims.

Claims

1. A cache system comprising: at least one DRAM array divided into a plurality of blocks; first and second SRAM arrays each having a capacity at least as large as the capacity of a block of the DRAM array; an address decoder coupled to receive a memory address and being operable to decode the address and generate decoded address signals corresponding thereto; an input/output circuit coupled to the DRAM array, the SRAM array, and the address decoder, the input/output circuit being operable to respond to a first control signal by coupling write data from an external data terminal to a location in a block of the DRAM array corresponding to the decoded address signals, or to respond to a second control signal by coupling write data from the external data terminal to a location in the first or second SRAM arrays, or to respond to a third control signal by coupling data from one of the SRAM arrays to a location in a block of the DRAM array corresponding to the decoded address signals; and a control circuit coupled to the DRAM array, the SRAM array, the address decoder, and the input/output circuit, the control circuit being operable to refresh the DRAM array one block at a time, the control circuit further being operable to generate the first control signal when the block of the DRAM array corresponding to the decoded address signals is not being refreshed, to generate the second control signal when the block of the DRAM array corresponding to the decoded address signals is being refreshed, to generate the third control signal when the block of the DRAM array that was being refreshed when the data was stored in the SRAM array is no longer being refreshed.
2. The cache system of claim 1 wherein the control circuit comprises: a refresh controller coupled to the DRAM array, the refresh controller being operable to refresh the DRAM array one block at a time; and a control buffer structured to receive a memory command and to generate the first, second and third control signals corresponding thereto.
3. The cache system of claim 1 wherein the block of DRAM array are physically part of a single DRAM array.
4. The cache system of claim 1 wherein the an input/output circuit comprises a plurality of complimentary input/output line pairs, the input/output lines being sufficient in number that a first pair of input/output lines may be coupled to one block of the DRAM array while a second pair of input/output lines may be coupled to another block of the DRAM array.
5. A cache system comprising: at least one DRAM array divided into a plurality of blocks; first and second SRAM arrays each having a capacity at least as large as the capacity of a block of the DRAM array, the SRAM arrays being coupled to the DRAM array for direct transfer of data from either of the SRAM arrays to any block of the DRAM array if the block of the DRAM array is not being refreshed; and an input/output circuit coupled to the DRAM array and the SRAM array, the input/output circuit being operable to store write data in a block of the DRAM array to which the write data was directed, and to store write data directed to any block of the DRAM array in either of the SRAM arrays if the block to which the write data was directed is being refreshed.
6. The cache system of claim 5 wherein the block of DRAM array are physically part of a single DRAM array.
7. The cache system of claim 5 wherein the an input/output circuit comprises a plurality of complimentary input/output line pairs, the input/output lines being sufficient in number that a first pair of input/output lines may be coupled to one block of the DRAM array while a second pair of input/output lines may be coupled to another block of the DRAM array.
8. A computer system, comprising: a processor having a processor bus; an input device coupled to the processor through the processor bus and adapted to allow data to be entered into the computer system; an output device coupled to the processor through the processor bus adapted to allow data to be output from the computer system; and a system controller coupled to the processor bus; a system memory coupled to the processor through the system controller; and a cache system coupled to the processor bus, the cache system comprising: at least one DRAM array divided into a plurality of blocks; first and second SRAM arrays each having a capacity at least as large as the capacity of a block of the DRAM array; a data transfer circuit coupled to the DRAM array and the SRAM array, the data transfer circuit being operable to respond to a first control signal by coupling write data to the first or second SRAM arrays, or to respond to a second control signal by coupling data from one of the SRAM arrays to a location in a block of the DRAM; and a control circuit coupled to the DRAM array and the SRAM array, the control circuit being operable to refresh the DRAM array one block at a time, the control circuit further being operable to generate the first control signal when the block of the DRAM array to which write data is being directed is being refreshed, and to generate the second control signal when the block of the DRAM array that was being refreshed when the data was stored in the SRAM array is no longer being refreshed.
9. The computer system of claim 8 wherein the control circuit comprises: a refresh controller coupled to the DRAM array, the refresh controller being operable to refresh the DRAM array one block at a time; and a control buffer structured to receive a memory command and to generate the first, second and third control signals corresponding thereto.
10. The computer system of claim 8 wherein the block of DRAM array are physically part of a single DRAM array.
11. The computer system of claim 8 wherein the. an input/output circuit comprises a plurality of complimentary input/output line pairs, the input/output lines being sufficient in number that a first pair of input/output lines may be coupled to one block of the DRAM array while a second pair of input/output lines may be coupled to another block of the DRAM array.
12. A computer system, comprising: a processor having a processor bus; an input device coupled to the processor through the processor bus and adapted to allow data to be entered into the computer system; an output device coupled to the processor through the processor bus adapted to allow data to be output from the computer system; and a system controller coupled to the processor bus; a system memory coupled to the processor through the system controller; and a cache system coupled to the processor bus, the cache system comprising: at least one DRAM array divided into a plurality of blocks; first and second SRAM arrays each having a capacity at least as large as the capacity of a block of the DRAM array; an input/output circuit coupled to the DRAM array and the SRAM array, the input/output circuit being operable to write data directed to any block of the DRAM array in either of the SRAM arrays if the block to which the write data was directed is being refreshed; and a control circuit coupled to the DRAM array and the SRAM array, the control circuit being operable to cause either of the SRAM, arrays to transfer data stored in the SRAM array to any block of the DRAM array if the block of the DRAM array is no longer being refreshed.
13. The computer system of claim 12 wherein the block of DRAM array are physically part of a single DRAM array.
14. The computer system of claim 12 wherein the an input/output circuit comprises a plurality of complimentary input/output line pairs, the input/output lines being sufficient in number that a first pair of input/output lines may be coupled to one block of the DRAM array while a second pair of input/output lines may be coupled to another block of the DRAM array.
15. A method of caching data, comprising: providing a DRAM having a plurality of blocks each of which may be refreshed; providing first and second SRAMs having a capacity of at least the capacity of the DRAM blocks; addressing a first block of the DRAM in an attempt to write data into the first block of the DRAM; if the first block of the DRAM is being refreshed during the attempt, writing the data into the first SRAM; after the first block of the DRAM is no longer being refreshed, transferring data stored in the first SRAM to the first block of the DRAM; addressing a second block of the DRAM in an attempt to write data into the second block of the DRAM; if the second block of the DRAM is being refreshed during the attempt, and data is being transferred from the first SRAM to the first block of the DRAM, writing the data into the second SRAM; and after the second block of the DRAM is no longer being refreshed, transferring data stored in the second SRAM to the second block of the DRAM.
16. The method of claim 15 further comprising writing data into the first SRAM if the second block of the DRAM is being refreshed during the attempt to write the data into the second block and data is not being transferred from the first SRAM to the first block of the DRAM.
17. The method of claim 15 further comprising writing the data into the first of the DRAM if the first block of the DRAM is not being refreshed during the attempt.
18. The method of claim 15 further comprising: addressing a third block of the DRAM in an attempt to write data into the third block of the DRAM; if the third block of the DRAM is being refreshed during the attempt, and data is being transferred from the second SRAM to the second block of the DRAM, writing the data into the first SRAM; and after the third block of the DRAM is no longer being refreshed, transferring data stored in the first SRAM to the third block of the DRAM.
19. The method of claim 15 further comprising: addressing a first block of the DRAM in an attempt to read data from the first block of the DRAM; and outputting data from the first block of the DRAM.
20. The method of claim 19 wherein the act of outputting data from the first block of the DRAM is accomplished regardless of whether or not the first block of memory is being refreshed.
21. The method of claim 15 further comprising: addressing the first block of the DRAM in an attempt to read data from the first block of the DRAM; and outputting data from the first block of the DRAM; and caching the data output from the first block of the DRAM in the first
SRAM.
22. The method of claim 21 further comprising addressing the first block of the DRAM in an attempt to read data from the first block of the DRAM; and outputting data from the first block of the first SRAM.
23. A method of caching data, comprising: providing a DRAM having a plurality of blocks each of which may be refreshed; providing first and second SRAMs having a capacity of at least the capacity of the DRAM blocks; sequentially attempting to write data to blocks of the DRAM; if a block to which data is attempting to be written is being refreshed, writing the data to one of the SRAMs; when the refresh of a block has been completed, transferring data from the SRAM to which the data had been written to a block of the DRAM to which data was attempted to be written; and if a block to which data is attempting to be written is being refreshed and data is being transferred from the one SRAM to a block of the DRAM, writing the data to the other SRAM.
24. The method of claim 23 further comprising, if the block to which data is attempting to be written is not being refreshed, writing the data to the block of the DRAM.
25. The method of claim 23 further comprising: attempting to read data from a block of the DRAM; and outputting data from the block of the DRAM to which an attempt to read data is directed.
26. The method of claim 25 wherein the act of outputting data from the block of the DRAM is accomplished regardless of whether or not the block of memory is being refreshed.
27. The method of claim 25 further comprising: attempting to read data from a block of the DRAM; outputting data from the block of the DRAM to which an attempt to read data is directed and caching the data output from the block of the DRAM in one of the SRAMs.
PCT/US2001/025494 2000-08-17 2001-08-14 Method and system for using dynamic random access memory as cache memory WO2002015019A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU2001284924A AU2001284924A1 (en) 2000-08-17 2001-08-14 Method and system for using dynamic random access memory as cache memory

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US09/642,546 2000-08-17
US09/642,546 US6862654B1 (en) 2000-08-17 2000-08-17 Method and system for using dynamic random access memory as cache memory

Publications (1)

Publication Number Publication Date
WO2002015019A1 true WO2002015019A1 (en) 2002-02-21

Family

ID=24577039

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2001/025494 WO2002015019A1 (en) 2000-08-17 2001-08-14 Method and system for using dynamic random access memory as cache memory

Country Status (3)

Country Link
US (5) US6862654B1 (en)
AU (1) AU2001284924A1 (en)
WO (1) WO2002015019A1 (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7123521B1 (en) 2005-04-27 2006-10-17 Micron Technology, Inc. Random cache read
US8601207B2 (en) * 2006-04-26 2013-12-03 The Invention Science Fund I, Llc Management of memory refresh power consumption
US9455035B2 (en) * 2006-04-26 2016-09-27 Invention Science Fund I, Llc Management of memory refresh power consumption
KR101975528B1 (en) 2012-07-17 2019-05-07 삼성전자주식회사 semiconductor memory cell array having fast array area and semiconductor memory including the same
US20140146589A1 (en) * 2012-11-29 2014-05-29 Samsung Electronics Co., Ltd. Semiconductor memory device with cache function in dram
US20140264915A1 (en) * 2013-03-15 2014-09-18 Chao-Yuan Huang Stacked Integrated Circuit System
US10402324B2 (en) * 2013-10-31 2019-09-03 Hewlett Packard Enterprise Development Lp Memory access for busy memory by receiving data from cache during said busy period and verifying said data utilizing cache hit bit or cache miss bit
KR102373544B1 (en) 2015-11-06 2022-03-11 삼성전자주식회사 Memory Device and Memory System Performing Request-based Refresh and Operating Method of Memory Device

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5991851A (en) * 1997-05-02 1999-11-23 Enhanced Memory Systems, Inc. Enhanced signal processing random access memory device utilizing a DRAM memory array integrated with an associated SRAM cache and internal refresh control
US6151269A (en) * 1992-11-06 2000-11-21 Mitsubishi Denki Kabushiki Kaisha Semiconductor memory device including dynamic type memory and static type memory formed on the common chip and an operating method thereof
US6170036B1 (en) * 1990-12-25 2001-01-02 Mitsubishi Denki Kabushiki Kaisha Semiconductor memory device and data transfer circuit for transferring data between a DRAM and a SRAM

Family Cites Families (53)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6071A (en) * 1849-01-30 Improved lubricating compound
US69325A (en) * 1867-10-01 photo-litho
JPS63247997A (en) 1987-04-01 1988-10-14 Mitsubishi Electric Corp Semiconductor storage device
CA2011518C (en) 1989-04-25 1993-04-20 Ronald N. Fortino Distributed cache dram chip and control method
JP2862948B2 (en) 1990-04-13 1999-03-03 三菱電機株式会社 Semiconductor storage device
US5359722A (en) 1990-07-23 1994-10-25 International Business Machines Corporation Method for shortening memory fetch time relative to memory store time and controlling recovery in a DRAM
US5414827A (en) 1991-12-19 1995-05-09 Opti, Inc. Automatic cache flush
EP0895162A3 (en) 1992-01-22 1999-11-10 Enhanced Memory Systems, Inc. Enhanced dram with embedded registers
US5471601A (en) 1992-06-17 1995-11-28 Intel Corporation Memory device and method for avoiding live lock of a DRAM with cache
US5473770A (en) 1993-03-02 1995-12-05 Tandem Computers Incorporated Fault-tolerant computer system with hidden local memory refresh
JP3305056B2 (en) 1993-08-31 2002-07-22 沖電気工業株式会社 Dynamic RAM
TW358907B (en) 1994-11-22 1999-05-21 Monolithic System Tech Inc A computer system and a method of using a DRAM array as a next level cache memory
US6128700A (en) 1995-05-17 2000-10-03 Monolithic System Technology, Inc. System utilizing a DRAM array as a next level cache memory and method for operating same
JPH0916470A (en) 1995-07-03 1997-01-17 Mitsubishi Electric Corp Semiconductor storage device
JPH0973776A (en) 1995-09-07 1997-03-18 Mitsubishi Electric Corp Synchronous semiconductor memory
US5835941A (en) 1995-11-17 1998-11-10 Micron Technology Inc. Internally cached static random access memory architecture
JP3352577B2 (en) 1995-12-21 2002-12-03 インターナショナル・ビジネス・マシーンズ・コーポレーション Storage device
US6131140A (en) 1995-12-22 2000-10-10 Cypress Semiconductor Corp. Integrated cache memory with system control logic and adaptation of RAM bus to a cache pinout
US5677878A (en) 1996-01-17 1997-10-14 Micron Technology, Inc. Method and apparatus for quickly restoring digit I/O lines
US6061759A (en) 1996-02-09 2000-05-09 Apex Semiconductor, Inc. Hidden precharge pseudo cache DRAM
US5796671A (en) 1996-03-01 1998-08-18 Wahlstrom; Sven E. Dynamic random access memory
US5875451A (en) 1996-03-14 1999-02-23 Enhanced Memory Systems, Inc. Computer hybrid memory including DRAM and EDRAM memory components, with secondary cache in EDRAM for DRAM
US6404670B2 (en) 1996-05-24 2002-06-11 Uniram Technology, Inc. Multiple ports memory-cell structure
UA42875C2 (en) * 1996-05-28 2001-11-15 Даймлеркрайслер Рейл Системз Гмбх Method of connecting the units of rail transportation means by gluing (variants)
US5813029A (en) 1996-07-09 1998-09-22 Micron Electronics, Inc. Upgradeable cache circuit using high speed multiplexer
TW349196B (en) 1996-10-18 1999-01-01 Ibm Cached synchronous DRAM architecture having a mode register programmable cache policy
US5787457A (en) 1996-10-18 1998-07-28 International Business Machines Corporation Cached synchronous DRAM architecture allowing concurrent DRAM operations
US5835401A (en) 1996-12-05 1998-11-10 Cypress Semiconductor Corporation Dram with hidden refresh
KR100242998B1 (en) 1996-12-30 2000-02-01 김영환 Structure of cell array and sense amplifier
US6088760A (en) 1997-03-07 2000-07-11 Mitsubishi Semiconductor America, Inc. Addressing system in a multi-port RAM having main and cache memories
US6172927B1 (en) * 1997-04-01 2001-01-09 Ramtron International Corporation First-in, first-out integrated circuit memory device incorporating a retransmit function
AU7706198A (en) 1997-05-30 1998-12-30 Micron Technology, Inc. 256 meg dynamic random access memory
US6094704A (en) 1997-06-17 2000-07-25 Micron Technology, Inc. Memory device with pipelined address path
US6023428A (en) 1997-07-28 2000-02-08 Texas Instruments Incorporated Integrated circuit device having a memory array with segmented bit lines and method of operation
JP3161383B2 (en) 1997-09-16 2001-04-25 日本電気株式会社 Semiconductor storage device
US6202119B1 (en) 1997-12-19 2001-03-13 Micron Technology, Inc. Method and system for processing pipelined memory commands
US6173356B1 (en) 1998-02-20 2001-01-09 Silicon Aquarius, Inc. Multi-port DRAM with integrated SRAM and systems and methods using the same
US6215497B1 (en) 1998-08-12 2001-04-10 Monolithic System Technology, Inc. Method and apparatus for maximizing the random access bandwidth of a multi-bank DRAM in a computer graphics system
US5999474A (en) * 1998-10-01 1999-12-07 Monolithic System Tech Inc Method and apparatus for complete hiding of the refresh of a semiconductor memory
US6415353B1 (en) 1998-10-01 2002-07-02 Monolithic System Technology, Inc. Read/write buffers for complete hiding of the refresh of a semiconductor memory and method of operating same
US6587918B1 (en) 1998-11-19 2003-07-01 Micron Technology, Inc. Method for controlling refresh of a multibank memory device
US6564284B2 (en) 1998-12-23 2003-05-13 Micron Technology, Inc. Apparatus for controlling a multibank memory device
US6172893B1 (en) 1999-01-05 2001-01-09 Micron Technology, Inc. DRAM with intermediate storage cache and separate read and write I/O
US6226755B1 (en) 1999-01-26 2001-05-01 Compaq Computer Corp. Apparatus and method for enhancing data transfer to or from a SDRAM system
US6178133B1 (en) 1999-03-01 2001-01-23 Micron Technology, Inc. Method and system for accessing rows in multiple memory banks within an integrated circuit
JP2000339954A (en) * 1999-05-31 2000-12-08 Fujitsu Ltd Semiconductor memory device
US6151664A (en) 1999-06-09 2000-11-21 International Business Machines Corporation Programmable SRAM and DRAM cache interface with preset access priorities
US6449690B1 (en) 1999-06-25 2002-09-10 Hewlett-Packard Company Caching method using cache data stored in dynamic RAM embedded in logic chip and cache tag stored in static RAM external to logic chip
US6445636B1 (en) 2000-08-17 2002-09-03 Micron Technology, Inc. Method and system for hiding refreshes in a dynamic random access memory
KR100702355B1 (en) 2000-08-30 2007-04-04 마이크론 테크놀로지, 인크 Semiconductor memory having dual port cell supporting hidden refresh
US6697909B1 (en) 2000-09-12 2004-02-24 International Business Machines Corporation Method and apparatus for performing data access and refresh operations in different sub-arrays of a DRAM cache memory
US6779076B1 (en) * 2000-10-05 2004-08-17 Micron Technology, Inc. Method and system for using dynamic random access memory as cache memory
US6629188B1 (en) 2000-11-13 2003-09-30 Nvidia Corporation Circuit and method for prefetching data for a texture cache

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6170036B1 (en) * 1990-12-25 2001-01-02 Mitsubishi Denki Kabushiki Kaisha Semiconductor memory device and data transfer circuit for transferring data between a DRAM and a SRAM
US6151269A (en) * 1992-11-06 2000-11-21 Mitsubishi Denki Kabushiki Kaisha Semiconductor memory device including dynamic type memory and static type memory formed on the common chip and an operating method thereof
US5991851A (en) * 1997-05-02 1999-11-23 Enhanced Memory Systems, Inc. Enhanced signal processing random access memory device utilizing a DRAM memory array integrated with an associated SRAM cache and internal refresh control

Also Published As

Publication number Publication date
US6948027B2 (en) 2005-09-20
AU2001284924A1 (en) 2002-02-25
US7917692B2 (en) 2011-03-29
US7155561B2 (en) 2006-12-26
US7350018B2 (en) 2008-03-25
US20060015679A1 (en) 2006-01-19
US20080177943A1 (en) 2008-07-24
US20070055818A1 (en) 2007-03-08
US6862654B1 (en) 2005-03-01
US20040186957A1 (en) 2004-09-23

Similar Documents

Publication Publication Date Title
US5226147A (en) Semiconductor memory device for simple cache system
US6041389A (en) Memory architecture using content addressable memory, and systems and methods using the same
US5226009A (en) Semiconductor memory device supporting cache and method of driving the same
US5179687A (en) Semiconductor memory device containing a cache and an operation method thereof
US5226139A (en) Semiconductor memory device with a built-in cache memory and operating method thereof
US7263020B2 (en) Memory device capable of refreshing data using buffer and refresh method thereof
US7917692B2 (en) Method and system for using dynamic random access memory as cache memory
US6445636B1 (en) Method and system for hiding refreshes in a dynamic random access memory
WO2000019445B1 (en) Read/write buffers for complete hiding of the refresh of a semiconductor memory and method of operating same
US6779074B2 (en) Memory device having different burst order addressing for read and write operations
US6779076B1 (en) Method and system for using dynamic random access memory as cache memory
US6282606B1 (en) Dynamic random access memories with hidden refresh and utilizing one-transistor, one-capacitor cells, systems and methods
US6400631B1 (en) Circuit, system and method for executing a refresh in an active memory bank
JP2006236550A (en) Integrated circuit device
JP2004288225A (en) Dram (dynamic random access memory) and access method
US7146454B1 (en) Hiding refresh in 1T-SRAM architecture
US5703810A (en) DRAM for texture mapping
KR100619202B1 (en) Apparatus and method for encoding auto-precharge
US6351419B1 (en) Integrated memory with a block writing function and global amplifiers requiring less space
US6064620A (en) Multi-array memory device, and associated method, having shared decoder circuitry
US10740188B2 (en) Volatile memory device and method for efficient bulk data movement, backup operation in the volatile memory device
US6433786B1 (en) Memory architecture for video graphics environment
US7333388B2 (en) Multi-port memory cells
US20230420027A1 (en) Memory device and refresh method thereof
JPH06243028A (en) Memory address control method

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT TZ UA UG UZ VN YU ZA ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): GH GM KE LS MW MZ SD SL SZ TZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

122 Ep: pct application non-entry in european phase
NENP Non-entry into the national phase

Ref country code: JP