US20040095796A1 - Multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and input/output assignments in an integrated circuit memory device - Google Patents
Multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and input/output assignments in an integrated circuit memory device Download PDFInfo
- Publication number
- US20040095796A1 US20040095796A1 US10/299,738 US29973802A US2004095796A1 US 20040095796 A1 US20040095796 A1 US 20040095796A1 US 29973802 A US29973802 A US 29973802A US 2004095796 A1 US2004095796 A1 US 2004095796A1
- Authority
- US
- United States
- Prior art keywords
- sub
- arrays
- blocks
- memory
- integrated circuit
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C5/00—Details of stores covered by group G11C11/00
- G11C5/02—Disposition of storage elements, e.g. in the form of a matrix array
- G11C5/025—Geometric lay-out considerations of storage- and peripheral-blocks in a semiconductor storage device
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C5/00—Details of stores covered by group G11C11/00
- G11C5/06—Arrangements for interconnecting storage elements electrically, e.g. by wiring
- G11C5/063—Voltage and signal distribution in integrated semi-conductor memory access lines, e.g. word-line, bit-line, cross-over resistance, propagation delay
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C8/00—Arrangements for selecting an address in a digital store
- G11C8/12—Group selection circuits, e.g. for memory block selection, chip selection, array selection
Definitions
- the present invention relates, in general, to the field of integrated circuit devices incorporating semiconductor memory components such as static random access memory (“SRAM”), dynamic random access memory (“DRAM”), electrically programmable read only memory (“EPROM”), ferroelectric random access memory (“FRAM”), Flash memory and the like. More particularly, the present invention relates to a multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and input/output (“I/O”) assignments in an integrated circuit memory device.
- SRAM static random access memory
- DRAM dynamic random access memory
- EPROM electrically programmable read only memory
- FRAM ferroelectric random access memory
- Flash memory Flash memory and the like. More particularly, the present invention relates to a multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and input/output (“I/O”) assignments in an integrated circuit memory device.
- the primary purpose of the memory device is to store data. Functionally, data may be written to the memory, read from it and, in the case of dynamic random access memory, periodically refreshed to maintain the integrity of the stored data.
- Each cell of a memory array serves to store a value or charge representative of either a logic level “1” or “0”.
- the memory cells are typically arranged into sub-arrays with row select circuitry on one side of the sub-arrays and data sensing circuitry on the two sides orthogonal to the row select circuitry.
- the overall memory is, in turn, built up into multiple banks of multiple sub-arrays in a uniform arrangement.
- such an arrangement serves to constrain the overall integrated circuit chip to then be of certain dimensions in the “X” and “Y” directions as determined by the size of the sub-arrays and the uniform arrangement of the sub-arrays into memory banks. While these constraints may not necessarily present a problem, there are instances where, for a given technology and package definition, they can present a problem.
- a multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and I/O assignments.
- non-uniform blocks of multiple identical sub-arrays non-uniform assignments of blocks to banks and/or nonuniform assignments of I/Os to blocks, it is possible to optimize the dimensions of the chip and the placement of the I/Os with respect to the package pads. In this manner, the granularity of the building blocks comprising multiple sub-arrays is improved while the flexibility in I/O assignment is also improved leading to more efficient and flexible chip layouts.
- an integrated circuit memory device which comprises a memory array having at least one memory bank.
- the memory bank comprises a plurality of blocks of sub-arrays wherein at least one of the plurality of blocks of sub-arrays is topologically non-uniform with respect to others of the blocks.
- the method comprises: partitioning the memory array into a plurality of blocks of sub-arrays with at least one of the blocks being topologically non-uniform with respect to others and arranging the plurality of blocks of sub-arrays such that data bit locations assigned to respective ones of the blocks of sub-arrays are substantially proximate to corresponding input/output lines of the memory array.
- an integrated circuit memory device comprising an array of memory cells formed into a plurality of memory banks.
- the memory device comprises at least one memory bank, with the memory bank further comprising a block of sub-arrays of the memory cells having a first on-chip area.
- At least one other memory bank comprises two or more additional blocks of sub-arrays of the memory cells, with each of the additional blocks of sub-arrays having a lesser on-chip area than the first on-chip area.
- FIG. 1 is a simplified block diagram of a conventional integrated circuit memory device including a memory array comprising four topologically uniform banks of memory cells arranged in uniform sub-arrays and indicating how a representative pair of I/Os must be routed to all four banks;
- FIG. 2 is a further simplified block diagram of a conventional integrated circuit memory device including a memory array made up of two separate subgroups of four uniform half-banks of memory cells and further indicating how each of a representative pair of I/Os may be routed to all of the memory half-banks in each of the sub-groups;
- FIG. 3 is a comparative block diagram of a representative integrated circuit memory device in accordance with the present invention wherein the architecture utilizes, with respect to at least one of the groups of memory banks, one or more topologically non-uniform blocks of sub-arrays and I/O assignments;
- FIG. 4A is a more detailed block diagram of a particular implementation of an integrated circuit memory device in accordance with the present invention for use, for example, in the implementation of an 8M ⁇ 32 (256 MB) device; and
- FIGS. 4B and 4C illustrate in greater detail the I/O assignment scheme for the particular implementation of a memory device in accordance with the representative embodiment of the present invention shown in the preceding figure.
- the memory device 100 includes a memory array 102 comprising four uniform banks 104 0 through 104 3 of memory cells arranged in uniform sub-arrays 114 .
- Each of the banks 104 include associated row select circuitry 106 and oppositely disposed data sensing circuitry 108 for accessing the sub-arrays thereof.
- a pair of data input/output (“I/O”) lines (or “DQ”) 110 and 112 are shown and each line (or groups of lines) must be routed to all of the banks 104 in the memory array 102 . Note that the entire row and column address fields are associated with each bank 104 .
- the memory cells are arranged in uniform groupings with the row select circuitry 106 on one side of the sub-arrays and the data sensing circuitry 108 on the two sides orthogonal to the row select circuitry 106 .
- the overall memory device 100 is, in turn, built up into multiple banks 104 of multiple sub-arrays in a uniform arrangement as shown with each bank 104 0 through 104 3 of the memory array 102 being identical to each of the other banks 104 .
- the memory device 200 includes a memory array 202 .
- the individual banks 0 - 3 are subdivided into two uniform groups of blocks of sub-arrays 204 , 206 .
- Each group includes one-half of banks 0 - 3 ( 208 0 - 208 3 in group 204 and 210 0 - 210 3 in group 206 ).
- the sub-division is done on the basis of the I/Os. That is, one group 204 is associated with I/O A 212 and the group 206 is associated with the I/O B 214 .
- the two sub-divisions are equal in the arrangement of the half-banks and the sub-arrays within the half-banks and in the silicon area that will be occupied by the two sub-divisions.
- any subdivision of the banks is always uniform and the specific I/Os are dedicated to specific sub-divisions of the banks.
- the entire row address field is associated with each half-bank in each sub-group.
- This prior art arrangement shown in FIGS. 1 and 2 constrain the overall integrated circuit chip to be of certain dimensions in the X and Y directions determined by the size of the sub-arrays and the uniform arrangement of the sub-arrays into banks 104 or half-banks 208 , 210 respectively.
- This constraint is normally not a problem, but for a given technology and package definition, it can become so given the size of the device and the design rules employed.
- the overall memory array is created by assembling multiple, uniform sub-arrays into uniform blocks by placing the sub-arrays side-by-side in a contiguous array.
- the sub-arrays within each block may all be assigned to a single bank or to multiple banks, but the format and physical dimensions of all the blocks are identical. Further, the distribution of the I/O assignments across the blocks is uniform, that is, there are the same number of blocks accessed for each I/O.
- FIG. 3 a comparative block diagram of a representative integrated circuit memory device 300 in accordance with the present invention.
- the illustrative memory device 300 shown includes a memory array 302 comprising a pair of sub-groups 304 and 306 , where each sub-group represents one-half of banks 0 - 3 .
- the sub-group 304 may be of a conventional design wherein each of the memory half-banks 308 0 through 308 3 are topologically uniform in configuration and include conventional I/O mapping.
- the sub-group 306 includes four memory half-banks 310 0 through 310 3 of which 310 0 and 310 , are uniform in the same manner as memory half-banks 308 0 through 308 3 of group 304 which are accessed by I/O A line 320 .
- at least one of the memory banks (in the illustrative embodiment shown, two memory banks) 310 2 and 310 3 is configured into topologically non-uniform blocks of sub-arrays with the former being formed into two sub-array blocks 312 and 314 while the latter is formed into two sub-array blocks 316 and 318 .
- the row address field of bank 2 and bank 3 are split between two sub-array blocks in contrast to the prior art in which the full row address filed is associated with all blocks of sub-arrays.
- the I/Os to the sub-arrays of group 306 may be advantageously arranged to access particular portions of certain of the banks 310 2 and 310 3 through I/O B line 322 .
- a given I/O line (or group of I/O lines) on the integrated circuit chip has, or could have, a bank/sub-array block arrangement different than that of the other I/O lines.
- I/O A line 320 may access the four banks 308 0 through 308 3 in a conventional “matched” situation while I/O B line 322 may be routed to its four bank 310 0 through 310 3 addresses by actually accessing five (or more) topologically non-uniform block destinations.
- the routing of I/O B line 322 for bank 2 ( 310 2 ) may be to either sub-array block 312 or the sub-array block 314 of bank 2 .
- access to bank 3 ( 310 3 ) may be to the sub-array block 318 or sub-array block 316 .
- the highest order row address may be used to determine which part of the two blocks 312 , 314 / 316 , 318 of 310 2 or 310 3 respectively should be accessed.
- FIG. 4A a more detailed block diagram of a particular implementation of an integrated circuit memory device 400 in accordance with the present invention for use, for example, in the implementation of an 8M ⁇ 32 (256 MB) device using 0.14 um design rules.
- the exemplary memory device 400 comprises, in pertinent part, a memory array 402 including four banks of memory cells denominated “0”, “1”, “2” and “3” comprising a number of topologically non-uniformly placed blocks of sub-arrays and wherein the data bit assignment numbers are indicated in the various sub-array blocks.
- the upper leftmost block of sub-arrays of bank “0” includes data bit numbers 12 - 15 and 24 - 27 out of 32 .
- Each of the blocks of sub-arrays of each of the banks “0”, “1”, “2” and “3” includes associated row select circuitry 404 and data sense circuitry (not shown).
- the memory device 400 further comprises a control block (“CONBLK”) 406 and an address/control block 408 .
- CONBLK control block
- the I/O lines to the various sub-arrays of the banks “0”, “1”, “2” and “3” are grouped into a number of regions, one of which is denominated DQ ⁇ 0:7> 410 , adjoining the lower left edge of the memory array 402 .
- DQ ⁇ 8:15> 412 adjoins the upper edge while DQ ⁇ 16:23> 414 is at the opposite lower edge.
- the I/O lines denominated DQ ⁇ 24:31> 416 adjoin the upper left edge of the memory array 402 .
- FIGS. 4B and 4C illustrate in greater detail the I/O assignment scheme for the particular implementation of a memory device 400 in accordance with the representative embodiment of present invention shown in the preceding figure.
- Corresponding structure to that illustrated in the preceding FIG. 4A is like numbered and the foregoing description thereof shall suffice herefor.
- the arrangement of the blocks of sub-arrays for the banks “0”, “1”, “2” and “3” are grouped in accordance with the method of the present invention.
- bank “0” has been arranged into five topologically non-uniform blocks of sub-arrays as shown.
- the banks “1”, “2” and “3”, while comprising topologically uniform blocks of sub-arrays, are nevertheless arranged in a topologically non-uniform fashion in order to also optimize placement of the I/o lines.
- the I/O lines have now been more effectively routed to the various blocks of sub-arrays as illustrated with the blocks containing the sub-arrays comprising data bit numbers 0 - 7 being advantageously positioned for access through DQ ⁇ 0:7> region 410 ; the data bit numbers 8 - 17 being advantageously positioned for access through DQ ⁇ 8:15> region 412 ; the data bit numbers 16 - 23 being advantageously positioned for access through DQ ⁇ 16:23> region 414 and the data bit numbers 24 - 31 being advantageously positioned for access through DQ ⁇ 24:31> region 416 .
- each of the I/O lines access two blocks of the sub-arrays of banks “1”, “2” and “3” but three blocks of the sub-arrays of bank “0”.
- the distribution of the I/O assignments across the blocks is non-uniform in that a differing number of blocks are actually accessed.
- the X and Y dimensions of the chip can be effectively optimized to meet package constraints for the memory device 400 without undue integrated circuit die size change or performance penalties.
Abstract
Description
- The present invention relates, in general, to the field of integrated circuit devices incorporating semiconductor memory components such as static random access memory (“SRAM”), dynamic random access memory (“DRAM”), electrically programmable read only memory (“EPROM”), ferroelectric random access memory (“FRAM”), Flash memory and the like. More particularly, the present invention relates to a multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and input/output (“I/O”) assignments in an integrated circuit memory device.
- Many types of relative large, commodity memory devices are currently available including those types listed above. Regardless of configuration, the primary purpose of the memory device is to store data. Functionally, data may be written to the memory, read from it and, in the case of dynamic random access memory, periodically refreshed to maintain the integrity of the stored data. Each cell of a memory array serves to store a value or charge representative of either a logic level “1” or “0”.
- In the design of semiconductor memories, the memory cells are typically arranged into sub-arrays with row select circuitry on one side of the sub-arrays and data sensing circuitry on the two sides orthogonal to the row select circuitry. The overall memory is, in turn, built up into multiple banks of multiple sub-arrays in a uniform arrangement. However, such an arrangement serves to constrain the overall integrated circuit chip to then be of certain dimensions in the “X” and “Y” directions as determined by the size of the sub-arrays and the uniform arrangement of the sub-arrays into memory banks. While these constraints may not necessarily present a problem, there are instances where, for a given technology and package definition, they can present a problem.
- A presentation and paper given at the 2002 International Solid State Circuits Conference (“ISSCC”) entitled: “The On-Chip 3MB Subarray Based 3rd Level Cache on an Itanium™ Microprocessor” by Don Weiss, John J. Wuu and Victor Chin describes a technique for utilizing multiple memory sub-arrays to build up the SRAM cache memory in a microprocessor through the use of a non-uniform sub-array placement. The so-called sub-arrays therein contemplated are specially designed, complete and relatively small SRAMs and each “sub-array” is dedicated to specific memory output bits, e.g. bits 0-7, with all of those specific bits (0-7) stored in the same “sub-array”.
- In contrast, with large commodity memory devices, it is simply not practical to have a single sub-array store all occurrences of specific memory output bits. In these types of devices, multiple sub-arrays are required. The design presented in the foregoing ISSCC description is intended for embedded microprocessor memory and is decidedly not a multi-bank design for use in packaged memory products with the associated size and pad placement constraints such designs impose.
- Disclosed herein is a multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and I/O assignments. By using non-uniform blocks of multiple identical sub-arrays, non-uniform assignments of blocks to banks and/or nonuniform assignments of I/Os to blocks, it is possible to optimize the dimensions of the chip and the placement of the I/Os with respect to the package pads. In this manner, the granularity of the building blocks comprising multiple sub-arrays is improved while the flexibility in I/O assignment is also improved leading to more efficient and flexible chip layouts.
- Particularly disclosed herein is an integrated circuit memory device which comprises a memory array having at least one memory bank. The memory bank, in turn, comprises a plurality of blocks of sub-arrays wherein at least one of the plurality of blocks of sub-arrays is topologically non-uniform with respect to others of the blocks.
- Further disclosed herein is a method for organizing a memory array layout in an integrated circuit device. The method comprises: partitioning the memory array into a plurality of blocks of sub-arrays with at least one of the blocks being topologically non-uniform with respect to others and arranging the plurality of blocks of sub-arrays such that data bit locations assigned to respective ones of the blocks of sub-arrays are substantially proximate to corresponding input/output lines of the memory array.
- Still further disclosed herein is an integrated circuit memory device comprising an array of memory cells formed into a plurality of memory banks. The memory device comprises at least one memory bank, with the memory bank further comprising a block of sub-arrays of the memory cells having a first on-chip area. At least one other memory bank comprises two or more additional blocks of sub-arrays of the memory cells, with each of the additional blocks of sub-arrays having a lesser on-chip area than the first on-chip area.
- The aforementioned and other features and objects of the present invention and the manner of attaining them will become more apparent and the invention itself will be best understood by reference to the following description of a preferred embodiment taken in conjunction with the accompanying drawings, wherein:
- FIG. 1 is a simplified block diagram of a conventional integrated circuit memory device including a memory array comprising four topologically uniform banks of memory cells arranged in uniform sub-arrays and indicating how a representative pair of I/Os must be routed to all four banks;
- FIG. 2 is a further simplified block diagram of a conventional integrated circuit memory device including a memory array made up of two separate subgroups of four uniform half-banks of memory cells and further indicating how each of a representative pair of I/Os may be routed to all of the memory half-banks in each of the sub-groups;
- FIG. 3 is a comparative block diagram of a representative integrated circuit memory device in accordance with the present invention wherein the architecture utilizes, with respect to at least one of the groups of memory banks, one or more topologically non-uniform blocks of sub-arrays and I/O assignments;
- FIG. 4A is a more detailed block diagram of a particular implementation of an integrated circuit memory device in accordance with the present invention for use, for example, in the implementation of an 8M×32 (256 MB) device; and
- FIGS. 4B and 4C illustrate in greater detail the I/O assignment scheme for the particular implementation of a memory device in accordance with the representative embodiment of the present invention shown in the preceding figure.
- With reference now to FIG. 1, a simplified block diagram of a conventional prior art integrated
circuit memory device 100 is shown. Thememory device 100 includes amemory array 102 comprising four uniform banks 104 0 through 104 3 of memory cells arranged inuniform sub-arrays 114. Each of the banks 104 include associated rowselect circuitry 106 and oppositely disposeddata sensing circuitry 108 for accessing the sub-arrays thereof. As illustrated, a pair of data input/output (“I/O”) lines (or “DQ”) 110 and 112 are shown and each line (or groups of lines) must be routed to all of the banks 104 in thememory array 102. Note that the entire row and column address fields are associated with each bank 104. - Stated another way, in the design of conventional
semiconductor memory devices 100, the memory cells are arranged in uniform groupings with the rowselect circuitry 106 on one side of the sub-arrays and thedata sensing circuitry 108 on the two sides orthogonal to the rowselect circuitry 106. Theoverall memory device 100 is, in turn, built up into multiple banks 104 of multiple sub-arrays in a uniform arrangement as shown with each bank 104 0 through 104 3 of thememory array 102 being identical to each of the other banks 104. - With reference additionally now to FIG. 2, a further simplified block diagram of another conventional prior art implementation of a
memory device 200 is shown. Thememory device 200 includes amemory array 202. The individual banks 0-3 are subdivided into two uniform groups of blocks ofsub-arrays group 204 and 210 0-210 3 in group 206). In the case of this implementation, the sub-division is done on the basis of the I/Os. That is, onegroup 204 is associated with I/O A 212 and thegroup 206 is associated with the I/O B 214. Further, the two sub-divisions are equal in the arrangement of the half-banks and the sub-arrays within the half-banks and in the silicon area that will be occupied by the two sub-divisions. In the prior art, any subdivision of the banks is always uniform and the specific I/Os are dedicated to specific sub-divisions of the banks. The entire row address field is associated with each half-bank in each sub-group. - This prior art arrangement shown in FIGS. 1 and 2 constrain the overall integrated circuit chip to be of certain dimensions in the X and Y directions determined by the size of the sub-arrays and the uniform arrangement of the sub-arrays into banks104 or half-
banks - In a conventional multi bank memory device100 (FIG. 1) or 200 (FIG. 2), the overall memory array is created by assembling multiple, uniform sub-arrays into uniform blocks by placing the sub-arrays side-by-side in a contiguous array. The sub-arrays within each block may all be assigned to a single bank or to multiple banks, but the format and physical dimensions of all the blocks are identical. Further, the distribution of the I/O assignments across the blocks is uniform, that is, there are the same number of blocks accessed for each I/O.
- With reference additionally now to FIG. 3, a comparative block diagram of a representative integrated
circuit memory device 300 in accordance with the present invention. Theillustrative memory device 300 shown includes amemory array 302 comprising a pair ofsub-groups sub-group 304 may be of a conventional design wherein each of the memory half-banks 308 0 through 308 3 are topologically uniform in configuration and include conventional I/O mapping. - As also shown, the
sub-group 306 includes four memory half-banks 310 0 through 310 3 of which 310 0 and 310, are uniform in the same manner as memory half-banks 308 0 through 308 3 ofgroup 304 which are accessed by I/O A line 320. In accordance with the present invention however, at least one of the memory banks (in the illustrative embodiment shown, two memory banks) 310 2 and 310 3 is configured into topologically non-uniform blocks of sub-arrays with the former being formed into twosub-array blocks sub-array blocks bank 2 andbank 3 are split between two sub-array blocks in contrast to the prior art in which the full row address filed is associated with all blocks of sub-arrays. In this manner, the I/Os to the sub-arrays ofgroup 306 may be advantageously arranged to access particular portions of certain of thebanks O B line 322. - In accordance with the technique of the present invention, a given I/O line (or group of I/O lines) on the integrated circuit chip has, or could have, a bank/sub-array block arrangement different than that of the other I/O lines. As illustrated, I/
O A line 320 may access the fourbanks 308 0 through 308 3 in a conventional “matched” situation while I/O B line 322 may be routed to its fourbank 310 0 through 310 3 addresses by actually accessing five (or more) topologically non-uniform block destinations. Forbanks 2 and 3 (310 2, 310 3), depending upon the row address, the routing of I/O B line 322 for bank 2 (310 2) may be to eithersub-array block 312 or thesub-array block 314 ofbank 2. Similarly, access to bank 3 (310 3), may be to thesub-array block 318 orsub-array block 316. In a representative implementation of the present invention, the highest order row address may be used to determine which part of the twoblocks - For a given I/
O line 322 within a bank, its particular path is defined when the row address is latched. Only the path of interest is enabled for the selectedbank 310 and the other path for the same bank is not enabled. By not requiring all thebanks 310 to be of equal block configuration and have equal I/O mapping, the integrated circuit die aspect ratio can be adjusted to fit any package size. - With reference additionally now to FIG. 4A, a more detailed block diagram of a particular implementation of an integrated
circuit memory device 400 in accordance with the present invention for use, for example, in the implementation of an 8M×32 (256 MB) device using 0.14 um design rules. Theexemplary memory device 400 comprises, in pertinent part, amemory array 402 including four banks of memory cells denominated “0”, “1”, “2” and “3” comprising a number of topologically non-uniformly placed blocks of sub-arrays and wherein the data bit assignment numbers are indicated in the various sub-array blocks. For example, the upper leftmost block of sub-arrays of bank “0” includes data bit numbers 12-15 and 24-27 out of 32. Each of the blocks of sub-arrays of each of the banks “0”, “1”, “2” and “3” includes associated rowselect circuitry 404 and data sense circuitry (not shown). - The
memory device 400 further comprises a control block (“CONBLK”) 406 and an address/control block 408. The I/O lines to the various sub-arrays of the banks “0”, “1”, “2” and “3” are grouped into a number of regions, one of which is denominated DQ<0:7> 410, adjoining the lower left edge of thememory array 402. DQ<8:15> 412 adjoins the upper edge while DQ<16:23> 414 is at the opposite lower edge. The I/O lines denominated DQ<24:31> 416 adjoin the upper left edge of thememory array 402. - With reference additionally now to FIGS. 4B and 4C illustrate in greater detail the I/O assignment scheme for the particular implementation of a
memory device 400 in accordance with the representative embodiment of present invention shown in the preceding figure. Corresponding structure to that illustrated in the preceding FIG. 4A is like numbered and the foregoing description thereof shall suffice herefor. - In these figures, the arrangement of the blocks of sub-arrays for the banks “0”, “1”, “2” and “3” are grouped in accordance with the method of the present invention. In particular, bank “0” has been arranged into five topologically non-uniform blocks of sub-arrays as shown. Further, it can be seen that the banks “1”, “2” and “3”, while comprising topologically uniform blocks of sub-arrays, are nevertheless arranged in a topologically non-uniform fashion in order to also optimize placement of the I/o lines. In particular, the I/O lines have now been more effectively routed to the various blocks of sub-arrays as illustrated with the blocks containing the sub-arrays comprising data bit numbers0-7 being advantageously positioned for access through DQ<0:7>
region 410; the data bit numbers 8-17 being advantageously positioned for access through DQ<8:15>region 412; the data bit numbers 16-23 being advantageously positioned for access through DQ<16:23>region 414 and the data bit numbers 24-31 being advantageously positioned for access through DQ<24:31>region 416. As indicated, each of the I/O lines access two blocks of the sub-arrays of banks “1”, “2” and “3” but three blocks of the sub-arrays of bank “0”. Thus, unlike in conventional devices, the distribution of the I/O assignments across the blocks is non-uniform in that a differing number of blocks are actually accessed. - In accordance with the method of the present invention, by not constraining all blocks of sub-arrays to be the same and by building the banks up out of non-uniform and non-contiguous assemblies, or blocks, of sub-arrays, the X and Y dimensions of the chip can be effectively optimized to meet package constraints for the
memory device 400 without undue integrated circuit die size change or performance penalties. - While there have been described above the principles of the present invention in conjunction with specific memory device types and architectures, it is to be clearly understood that the foregoing description is made only by way of example and not as a limitation to the scope of the invention. Particularly, it is recognized that the teachings of the foregoing disclosure will suggest other modifications to those persons skilled in the relevant art. Such modifications may involve other features which are already known per se and which may be used instead of or in addition to features already described herein. Although claims have been formulated in this application to particular combinations of features, it should be understood that the scope of the disclosure herein also includes any novel feature or any novel combination of features disclosed either explicitly or implicitly or any generalization or modification thereof which would be apparent to persons skilled in the relevant art, whether or not such relates to the same invention as presently claimed in any claim and whether or not it mitigates any or all of the same technical problems as confronted by the present invention. The applicants hereby reserve the right to formulate new claims to such features and/or combinations of such features during the prosecution of the present application or of any further application derived therefrom.
Claims (22)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/299,738 US6741488B1 (en) | 2002-11-19 | 2002-11-19 | Multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and input/output assignments in an integrated circuit memory device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/299,738 US6741488B1 (en) | 2002-11-19 | 2002-11-19 | Multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and input/output assignments in an integrated circuit memory device |
Publications (2)
Publication Number | Publication Date |
---|---|
US20040095796A1 true US20040095796A1 (en) | 2004-05-20 |
US6741488B1 US6741488B1 (en) | 2004-05-25 |
Family
ID=32297766
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/299,738 Expired - Lifetime US6741488B1 (en) | 2002-11-19 | 2002-11-19 | Multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and input/output assignments in an integrated circuit memory device |
Country Status (1)
Country | Link |
---|---|
US (1) | US6741488B1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7426566B2 (en) | 2001-01-17 | 2008-09-16 | International Business Machines Corporation | Methods, systems and computer program products for security processing inbound communications in a cluster computing environment |
US11327838B2 (en) * | 2018-04-20 | 2022-05-10 | Samsung Electronics Co., Ltd. | Memory device having error correction function and operating method thereof |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003085672A1 (en) * | 2002-04-10 | 2003-10-16 | Hynix Semiconductor Inc. | Memory chip architecture having non-rectangular memory banks and method for arranging memory banks |
US20060161743A1 (en) * | 2005-01-18 | 2006-07-20 | Khaled Fekih-Romdhane | Intelligent memory array switching logic |
US7646622B2 (en) * | 2006-03-23 | 2010-01-12 | Toshiba America Research, Inc. | Memory based computation systems and methods of using the same |
JP2009004042A (en) * | 2007-06-22 | 2009-01-08 | Nec Electronics Corp | Semiconductor memory device |
US8120985B2 (en) * | 2008-03-12 | 2012-02-21 | Qimonda Ag | Multi-bank memory device method and apparatus |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6173356B1 (en) * | 1998-02-20 | 2001-01-09 | Silicon Aquarius, Inc. | Multi-port DRAM with integrated SRAM and systems and methods using the same |
US6256256B1 (en) * | 1998-01-30 | 2001-07-03 | Silicon Aquarius, Inc. | Dual port random access memories and systems using the same |
-
2002
- 2002-11-19 US US10/299,738 patent/US6741488B1/en not_active Expired - Lifetime
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6256256B1 (en) * | 1998-01-30 | 2001-07-03 | Silicon Aquarius, Inc. | Dual port random access memories and systems using the same |
US6173356B1 (en) * | 1998-02-20 | 2001-01-09 | Silicon Aquarius, Inc. | Multi-port DRAM with integrated SRAM and systems and methods using the same |
US6504785B1 (en) * | 1998-02-20 | 2003-01-07 | Silicon Aquarius, Inc. | Multiprocessor system with integrated memory |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7426566B2 (en) | 2001-01-17 | 2008-09-16 | International Business Machines Corporation | Methods, systems and computer program products for security processing inbound communications in a cluster computing environment |
US11327838B2 (en) * | 2018-04-20 | 2022-05-10 | Samsung Electronics Co., Ltd. | Memory device having error correction function and operating method thereof |
Also Published As
Publication number | Publication date |
---|---|
US6741488B1 (en) | 2004-05-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111788685B (en) | Three-dimensional memory device | |
KR100268961B1 (en) | Semiconductor device and method for fabricating the same, memory core chip and memory peripheral circuit chip | |
US8120985B2 (en) | Multi-bank memory device method and apparatus | |
US20030012045A1 (en) | Architecture for a semiconductor memory device | |
US5150330A (en) | Interblock dispersed-word memory architecture | |
JPH0240193A (en) | Dynamic semiconductor memory | |
US20070162685A1 (en) | Memory data bus structure and method of transferring information with plural memory banks | |
US9978438B2 (en) | Device having multiple switching buffers for data paths controlled based on IO configuration modes | |
US6448602B1 (en) | Semiconductor memory device with improved arrangement of memory blocks and peripheral circuits | |
US6741488B1 (en) | Multi-bank memory array architecture utilizing topologically non-uniform blocks of sub-arrays and input/output assignments in an integrated circuit memory device | |
US5812418A (en) | Cache sub-array method and apparatus for use in microprocessor integrated circuits | |
US5293334A (en) | Pattern layout of power source lines in semiconductor memory device | |
KR20100083626A (en) | Semiconductor memory device | |
US6925543B2 (en) | Burst transfer memory | |
US4594690A (en) | Digital storage apparatus including sections exhibiting different access speeds | |
KR20040111368A (en) | Nonvolatile semiconductor storage device | |
CN113454719B (en) | Centralized placement of commands and addresses in a memory device | |
US6169699B1 (en) | Semiconductor memory device | |
KR100762867B1 (en) | Semiconductor memory device with global io line | |
US20180102161A1 (en) | Vertical Thyristor Memory Array and Memory Array Tile Therefor | |
US6215718B1 (en) | Architecture for large capacity high-speed random access memory | |
US7038932B1 (en) | High reliability area efficient non-volatile configuration data storage for ferroelectric memories | |
KR100314129B1 (en) | Semiconductor implementing bank and data input/output line architecture to reduce data input/output line loading | |
US11961585B2 (en) | Centralized placement of command and address swapping in memory devices | |
KR100206704B1 (en) | Method of forming the array of semiconductor memory device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: MOSEL VITELIC INC., TAIWAN Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:HEIGHTLEY, JOHN;FAUE, JON ALLAN;REEL/FRAME:013512/0526 Effective date: 20021115 |
|
AS | Assignment |
Owner name: PROMOS TECHNOLOGIES INC., TAIWAN Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:MOSEL VITELIC INC.;REEL/FRAME:014326/0524 Effective date: 20040202 |
|
AS | Assignment |
Owner name: PROMOS TECHNOLOGIES INC., TAIWAN Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:MOSEL VITELIC INC.;REEL/FRAME:014391/0577 Effective date: 20040202 |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
CC | Certificate of correction | ||
FPAY | Fee payment |
Year of fee payment: 4 |
|
AS | Assignment |
Owner name: SAMSUNG ELECTRONICS CO., LTD., KOREA, REPUBLIC OF Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:PROMOS TECHNOLOGIES;REEL/FRAME:023758/0289 Effective date: 20090811 |
|
FPAY | Fee payment |
Year of fee payment: 8 |
|
FPAY | Fee payment |
Year of fee payment: 12 |