Classifications

• • G—PHYSICS
  • G11—INFORMATION STORAGE
  • G11C—STATIC STORES
  • G11C16/00—Erasable programmable read-only memories
  • G11C16/02—Erasable programmable read-only memories electrically programmable
  • G11C16/06—Auxiliary circuits, e.g. for writing into memory
  • G11C16/10—Programming or data input circuits
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; 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
  • G06F12/0855—Overlapped cache accessing, e.g. pipeline
• • 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
  • G11C11/56—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using storage elements with more than two stable states represented by steps, e.g. of voltage, current, phase, frequency
  • G11C11/5621—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using storage elements with more than two stable states represented by steps, e.g. of voltage, current, phase, frequency using charge storage in a floating gate
  • G11C11/5628—Programming or writing circuits; Data input circuits
• • 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
  • G11C11/56—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using storage elements with more than two stable states represented by steps, e.g. of voltage, current, phase, frequency
  • G11C11/5621—Digital stores characterised by the use of particular electric or magnetic storage elements; Storage elements therefor using storage elements with more than two stable states represented by steps, e.g. of voltage, current, phase, frequency using charge storage in a floating gate
  • G11C11/5642—Sensing or reading circuits; Data output circuits
• • 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
  • G11C7/1039—Read-write modes for single port memories, i.e. having either a random port or a serial port using pipelining techniques, i.e. using latches between functional memory parts, e.g. row/column decoders, I/O buffers, sense amplifiers
• • 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
  • G11C7/1078—Data input circuits, e.g. write amplifiers, data input buffers, data input registers, data input level conversion circuits
  • G11C7/1087—Data input latches
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
  • G06F2212/20—Employing a main memory using a specific memory technology
  • G06F2212/202—Non-volatile memory
  • G06F2212/2022—Flash memory
• • G—PHYSICS
  • G11—INFORMATION STORAGE
  • G11C—STATIC STORES
  • G11C2207/00—Indexing scheme relating to arrangements for writing information into, or reading information out from, a digital store
  • G11C2207/22—Control and timing of internal memory operations
  • G11C2207/2245—Memory devices with an internal cache buffer
• • G—PHYSICS
  • G11—INFORMATION STORAGE
  • G11C—STATIC STORES
  • G11C2211/00—Indexing scheme relating to digital stores characterized by the use of particular electric or magnetic storage elements; Storage elements therefor
  • G11C2211/56—Indexing scheme relating to G11C11/56 and sub-groups for features not covered by these groups
  • G11C2211/562—Multilevel memory programming aspects
  • G11C2211/5623—Concurrent multilevel programming and reading
• • G—PHYSICS
  • G11—INFORMATION STORAGE
  • G11C—STATIC STORES
  • G11C2211/00—Indexing scheme relating to digital stores characterized by the use of particular electric or magnetic storage elements; Storage elements therefor
  • G11C2211/56—Indexing scheme relating to G11C11/56 and sub-groups for features not covered by these groups
  • G11C2211/564—Miscellaneous aspects
  • G11C2211/5642—Multilevel memory with buffers, latches, registers at input or output
• • G—PHYSICS
  • G11—INFORMATION STORAGE
  • G11C—STATIC STORES
  • G11C2211/00—Indexing scheme relating to digital stores characterized by the use of particular electric or magnetic storage elements; Storage elements therefor
  • G11C2211/56—Indexing scheme relating to G11C11/56 and sub-groups for features not covered by these groups
  • G11C2211/564—Miscellaneous aspects
  • G11C2211/5643—Multilevel memory comprising cache storage devices
• • 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
  • G11C7/1078—Data input circuits, e.g. write amplifiers, data input buffers, data input registers, data input level conversion circuits

Definitions

• This invention relates generally to non- volatile semiconductor memory such as electrically erasable programmable read-only memory (EEPROM) and flash EEPROM, and specifically to cache operations based on shared latch structures allowing overlapping memory operations.
• EEPROM electrically erasable programmable read-only memory
• flash EEPROM flash EEPROM
• non-volatile memories such as flash memory
• flash memory with multiple levels per cell can be used to replace the binary chips; however, the speed of operations can be slower in multi-state memories, such as in the case of writing data where the tolerances between states become stricter. Consequently, the performance level of memories having multi-level cells has much scope for improvement.
• a non-volatile memory and corresponding method of operation are presented, where the memory has addressable pages of memory cells and each memory cell of an addressed page is provided with a set of corresponding data latches that can latch a predetermined number of bits.
• the memory can perform a first operation (such as a write, for example) on a designated group of one or more addressed pages using a first set of data stored in the corresponding set of data latches and also receive a request for a second operation (such as a read, for example) that also uses some of these corresponding data latches with a second set of data.
• the memory subsequently performs the second operation during the first operation when a sufficient number of latches become available; and if, in response to determining whether there are a sufficient number of the corresponding set of data latches to perform the second operation it is determined that there are a sufficient number, performing the second operation during the first operation.
• FIG. 1 illustrates inserting a read inside a cache program operation.
• FIG. 2 illustrates inserting a read inside a cache erase operation.
• FIG. 3 shows a particular programming order for pages and some corresponding look ahead reads.
• FIG. 4 illustrates inserting a read inside a cache program operation when there are insufficient latches available.
• FIG. 5 illustrates inserting a read inside a cache erase operation when there are insufficient latches available.
• FIG. 6 is another example of inserting a read inside a cache program operation when there are insufficient latches available.
• FIG. 7 shows schematically how various cache points occur.
• FIG. 8 is a flow chart for one basic embodiment of the adaptive algorithm.
• Non- volatile memories such as NAND flash memory
• NAND flash memory with multi levels per cell are used to replace binary chips
• One set of performance enhancements is based on utilizing complicated cache algorithm to do parallel operations at the same time. Examples of some such enhancements are given US patent application publication numbers US- 2006-0221704-A1 and US-2007-0109867-A1. Reference is also made to the following US patent application publication numbers, which provide additional detail on many aspects: US 2006-0233023-A1; US 2006-0233021-A1; US 2006-0221696- Al; US 2006-0233010-Al; US 2006-0239080-A1; and US 2007-0002626-A1.
• Look Ahead read is an algorithm that uses a corrective reading that depends on the data of the next word.
• Read with the LA (“Look Ahead”) correction basically examines the memory states programmed into the cells on an adjacent wordline and corrects any perturbation effect they have on the memory cells being read on the current wordline. If the pages have been programming according to a preferred programming scheme described in the cited references, then the adjacent wordline will be from the wordline immediately above the current one. The LA correction scheme would require the data on the adjacent wordline to be read prior to the current page.
• the number of data latches required to do a LA read will depend on the correction that is required. In some cases this will be a 1 bit correction, others will use a 2 bit or 3 bit correction. The correction needed for each page will depend on the program sequence that the page and the adjacent pages went through. In some cases, one page will need only 1 bit correction, while another page will possibly use 2 bit correction. These different correction levels will use different numbers of data latches to handle the LA read.
• mixed cache operations such as inserting a read in cache program for a copy function, or inserting a read in a cache erase operation
• the variations of data latch requirements for the read is preferably accounted for in the cache algorithms.
• the data latch requirement is also unknown before the user (e.g., a controller or host) issues the address. To better handle these complications a new cache algorithm, called adaptive algorithm in the following, is introduced.
• LA look ahead
• LM lower middle
• Read with the LA ("Look Ahead”) correction basically examines the memory states programmed into the cells on an adjacent wordline and corrects any perturbation effect they have on the memory cells being read on the current wordline. If the pages have been programming according to the preferred programming scheme described above, then the adjacent wordline will be from the wordline immediately above the current one. The LA correction scheme would require the data on the adjacent wordline to be read prior to the current page.
• Figures 1 and 2 respectively give example of inserting a read into a cache program and a cache erase operation. In the cache program with copy operation of Figure 1, the process begins with the program operation beginning at 101.
• Figure 2 is the corresponding arrangement for a cache erase with read.
• An erase process here with including a soft programming operation, is started at 210.
• data latches are avail for an interposed operation. Since a soft program operation can be considered a sort of binary programming operation, for N-state memory cells, this will typically result in the there being (N-I) available latches.
• a read operation can then be interposed at 205, after which the soft programming phase can continue.
• the insert read operation (105 or 111 in Figure 1, 205 in Figure 2) is a look ahead read operation
• the data latch requirements will depend on the amount of correction used.
• 1 bit correction will use 2 data latches, with one data latch is for WLn+ 1 data and one for 1 page of WLn data.
• 3 data latches are used (two for WLn+ 1 data and one for 1 page of WLn data)
• 4 data latches are used (three for WLn+ 1 data and one for 1 page of WLn data).
• the inclusion of the lower middle (“LM”) page order and the corresponding latch requirements are consider when combined with the LA read, an arrangement that is developed in more detail in a US patent application entitled “Different Combinations of Wordline Order and Look-Ahead read to improve Non- Volatile Memory performance" of Yan Li, filed March 19, 2008.
• the pages may be arranged so that the lower and middle are consecutive and programmed together, but where the upper page will be jumped in a way that upper page program will tend to eliminate the middle page to middle page WL-WL coupling effects.
• the upper page is programmed after the next wordline's middle page program.
• the lower and middle page will be read in with LA read to correctively read in data from the memory cells.
• an upper page read will need only 1 bit correction, since the upper page is only coupled to the next WL upper page program.
• a middle page read on the other hand, will use 2 bits for LA read correction, since lower and middle page can couple the previous wordlines middle page voltage thresholds.
• FIG. 4 Going back to the example of a cache program with an inserted read for copy case, this is shown in Figure 4, where the program operation begins at 401. As the operation continues, at some point a pair of latches for each cell in the page is freed up at 403. At this point a read can be inserted as shown at 405. This may either be a read that the state machine was already holding, waiting for latches to open up, or a read request that comes in after the latches open, in which case the programming would have continued until the read came in. In either case, once the read is inserted, it may be determined that 2 latches are not enough to complete this read.
• a read command may be entered when 2 data latch is available, but may only be executed by assuming that the upper page read uses only a 1 bit LA read; however, if the page does not have its upper page programmed, it will need 2 bits for the LA read. Typically, the latches will have been filled with program data, but once it is determined that further latches are needed, the read data would be treated as invalid. In this situation, the read command the user (i.e., the controller) issued cannot be completed until more data latches are available. In this example, the read cannot be executed until 3 data latches are available. Once the memory determines that there are insufficient latches and the read cannot be completed, the command held until the needed latches are available, as shown by the arrow.
• the erase with soft program begins at 501.
• Figure 6 illustrates another situation, where the adaptive cache operation could be complicated to manage.
• Figure 6 shows an example where a first read is executed in 601-607 without needing more latches than the available latches, much as in patent publications referenced above. This read-in page will be checked for ECC and then ready to be programmed into another location. During the upper page program (part of 607), a second read can be inserted at 611. If the second read can not be executed due to the unavailability of sufficient data latches, the second read can not be executed immediately and will wait until the upper page finish its program (613).
• the first read data (which has not yet been programmed) is still in the data latches will be transferred to the right location.
• the second read command can be executed after the upper page program finished.
• the incomplete read can be executed again (617), much as described with respect to Figure 4.
• FIG. 7 shows this conceptually, where time is taken as running to the right and diagram picks up at some more or less arbitrary point in operation of an ongoing write command.
• the various available cache points are as indicated at 701, 703, ..., 713, where the write operation that was in process when Figure 7 begins ends at 731 , after which the next write takes up.
• FIG. 8 is a flowchart of one exemplary embodiment.
• a command for the operation will be issued and at a cache point (indicated by a Ready/Busy signal, for example) the command is entered.
• the state machine After the user command issued, the state machine will check if there are enough data latches to execute this command at 803. If there are enough data latches (yes out of 805), then the user command can be executed immediately at 807 and then return to 801.

Landscapes

• Engineering & Computer Science (AREA)
• Theoretical Computer Science (AREA)
• Computer Hardware Design (AREA)
• Physics & Mathematics (AREA)
• General Engineering & Computer Science (AREA)
• General Physics & Mathematics (AREA)
• Read Only Memory (AREA)
• Memory System Of A Hierarchy Structure (AREA)

Priority Applications (2)

Applications Claiming Priority (2)

Publications (1)

Family

ID=40547403

Family Applications (1)

Country Status (6)

Families Citing this family (13)

Citations (3)

Family Cites Families (10)

• 2008
  • 2008-03-19 US US12/051,462 patent/US7961512B2/en active Active
• 2009
  • 2009-02-19 CN CN2009801179088A patent/CN102037519A/zh active Pending
  • 2009-02-19 WO PCT/US2009/034573 patent/WO2009117204A1/en not_active Ceased
  • 2009-02-19 KR KR1020107020949A patent/KR20100138943A/ko not_active Ceased
  • 2009-02-19 JP JP2011500833A patent/JP2011515786A/ja active Pending
  • 2009-03-09 TW TW098107583A patent/TWI382310B/zh not_active IP Right Cessation

Patent Citations (3)

Also Published As

Similar Documents

Legal Events