CN107239409A - A kind of significant data distribution method and its system based on temperature - Google Patents
A kind of significant data distribution method and its system based on temperature Download PDFInfo
- Publication number
- CN107239409A CN107239409A CN201710316728.9A CN201710316728A CN107239409A CN 107239409 A CN107239409 A CN 107239409A CN 201710316728 A CN201710316728 A CN 201710316728A CN 107239409 A CN107239409 A CN 107239409A
- Authority
- CN
- China
- Prior art keywords
- temperature
- minimum
- sample point
- subregion
- layer
- 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
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
- G06F12/0223—User address space allocation, e.g. contiguous or non contiguous base addressing
- G06F12/023—Free address space management
- G06F12/0238—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
- G06F12/0246—Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory in block erasable memory, e.g. flash memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3058—Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Quality & Reliability (AREA)
- Read Only Memory (AREA)
Abstract
The present invention provides a kind of significant data distribution method based on temperature, applied to three-dimensional flash memory, wherein, methods described includes：When the operating system of the three-dimensional flash memory sends the request for writing default significant data, the temperature of chip each several part in the three-dimensional flash memory is monitored；The default method of salary distribution is selected to carry out default significant data write operation according to the default cooling model of the uniformity and operating system that monitor chip each several part Temperature Distribution.The present invention also provides a kind of significant data distribution system based on temperature.The technical scheme that the present invention is provided according to uniformity of temperature profile whether and cooling model it is whether powerful, there is provided 4 kinds of different block allocation strategies, the physical block that ensure that is used to store under different scenes has higher stability and reliability, prevents significant data from producing mistake.
Description
Technical field
The present invention relates to field of data storage, more particularly to a kind of significant data distribution method based on temperature and its it is
System.
Background technology
3D flash memories (Three-dimensional flash memory) are a kind of emerging cutting edge technologies, with splendid
Development prospect.It breaches conventional two-dimensional planar flash memory (2D/planar flash memory) bottleneck, in guaranteed capacity and
On the premise of reliability, it is not necessary to expend big cost and effort again and go to improve making technology, but consider to stack more numbers of plies
To realize.Plane nand flash memory uses the structure of individual layer (Single layer), and the size of memory cell is with making technology
Improve and progressively reduce, the density of memory cells increase in unit area, relatively the interference between unit also will be increasingly
Seriously, so that causing the reliability of nand flash memory product reduces.
3D flash memories use the stereochemical structure of multilayer (Multi-layer), and each plane layer is stacked.For flash chip
Single plane for, the limited amount system of superincumbent memory cell can be placed, which limits the overall storage of equipment
Capacity.And according to the design feature of 3D flash memories, if dried layer can be classified as, every layer has some rows, often row has several physical blocks.
Based on such stereochemical structure, 3D flash memories allow memory cell vertical stacking so that flush memory device capacity can be exponentially
Increase, realizes higher capacity in smaller physical space.
In the technology of existing physical space distribution, such as " interference mitigation allocative decision " (Y.-M.Chang, Y.-
H.Chang,T.-W.Kuo,H.-P.Li,and Y.-C.Li,“A disturb-alleviation scheme for 3D
flash memory,”in 2013IEEE/ACM International Conference on Computer-Aided
Design (ICCAD), Nov 2013, pp.421-428) and " the enhanced address of cache strategy of reliability " (Y.Wang, Z.Shao,
H.Chan,L.Bathen,and N.Dutt,“A reliability enhanced address mapping strategy
for three-dimensional(3-D)NAND flash memory,”IEEE Transactions on Very Large
Scale Integration (VLSI) Systems, vol.22, no.11, pp.2402-2410, Nov 2014), do not consider
To influence of the temperature to chip.The mapping table in actual address space takes the side of Sequential Mapping in physical space and flash chip
Formula, and temperature has proximity, can thus there is the problem of contiguous physical block is interfered.
For the physical block batch operation of significant data, in existing 3D flash memory managements technology, do not examine well
Consider influence of the temperature to flash memory.3D flash chips are very sensitive to temperature.In fact, the radiation and shifting of high temperature meeting accelerated charge
It is dynamic, cause magnitude of voltage on physical block to change and (such as become by 01 and turn to 11), this allows for data storage and malfunctioned.For class
For the significant data for being similar to file system metadata and the class of address mapping table one, such mistake is fatal.If even literary
The address of part system can not all be found, then can not possibly more go load document system.So, write for such significant data
Enter the distribution of block, it is desirable to high stability and high integrality.But, the data stability of currently available technology in this respect and complete
Property is generally relatively low.
The content of the invention
In view of this, it is an object of the invention to provide a kind of significant data distribution method and its system based on temperature,
Exist when aiming to solve the problem that physical block batch operation for significant data in the prior art significant data stability and integrality compared with
Low the problem of.
The present invention proposes a kind of significant data distribution method based on temperature, applied to three-dimensional flash memory, wherein, methods described
Including：
When the operating system of the three-dimensional flash memory sends the request for writing default significant data, monitor in the three-dimensional flash memory
The temperature of chip each several part；
It is pre- according to the default cooling model selection of the uniformity and operating system that monitor chip each several part Temperature Distribution
If the method for salary distribution carry out default significant data write operation.
On the other hand, the present invention also provides a kind of significant data distribution system based on temperature, applied to three-dimensional flash memory, its
It is characterised by, the system includes：
Temperature measurement module, for when the operating system of the three-dimensional flash memory sends the request for writing default significant data, monitoring
The temperature of chip each several part in the three-dimensional flash memory；
Distribute module, for according to the default drop of uniformity and operating system for monitoring chip each several part Temperature Distribution
Temperature model selects the default method of salary distribution to carry out default significant data write operation.
Technical scheme temperature in use sensor or the temperature of thermal sensation camera dynamic monitoring 3D flash chips that the present invention is provided,
Default significant data is assigned to the low block region of temperature, to ensure the integrality and stability of default significant data.The present invention
The technical scheme of offer according to uniformity of temperature profile whether and cooling model it is whether powerful, build 4 kinds of different blocks distribution plans
Slightly, it is ensured that the physical block for being used to store under different scenes has higher stability, prevents significant data from producing mistake, makes simultaneously
Aid in realizing the Search and Orientation of block with quaternary tree and queue data structure, the low block region of temperature can be navigated to, and keep away
Exempt to perform the situation of write operation repeatedly in same area, reduced the accumulation of heat, improve abrasion equilibrium.
Brief description of the drawings
Fig. 1 is the significant data distribution method flow chart based on temperature in an embodiment of the present invention；
Fig. 2 is that Fig. 1 is monitoring that chip each several part temperature is uniformly distributed and default cooling in an embodiment of the present invention
Default significant data distribution method flow chart when model is strong cooling model；
Fig. 3 is that Fig. 1 is monitoring that chip each several part temperature is uniformly distributed and default cooling in an embodiment of the present invention
Default significant data distribution method flow chart when model is weak cooling model；
Fig. 4 is monitoring chip each several part temperature non-uniform Distribution and default drop for Fig. 1 in an embodiment of the present invention
Default significant data distribution method flow chart when temperature model is strong cooling model；
Fig. 5 is monitoring chip each several part temperature non-uniform Distribution and default drop for Fig. 1 in an embodiment of the present invention
Default significant data distribution method flow chart when temperature model is weak cooling model；
Fig. 6 is the internal structure schematic diagram of the significant data distribution system 10 based on temperature in an embodiment of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
A kind of significant data distribution method based on temperature provided by the present invention will be described in detail below.
Referring to Fig. 1, being the significant data distribution method flow chart based on temperature in an embodiment of the present invention.
In the present embodiment, the significant data distribution method based on temperature is applied to three-dimensional flash memory (Three-
Dimensional flash memory), the significant data includes default operating system metadata, address mapping table, bag
Data for containing specific and exclusive data, frequently accessing or updating etc..
In step sl, when the operating system of the three-dimensional flash memory sends the request for writing the default significant data, prison
Survey the temperature of chip each several part in the three-dimensional flash memory.
In the present embodiment, when operating system sends the request for writing the default significant data, important number is indicated
According to (for example：File system metadata, address mapping table etc.) need to write three-dimensional flash memory chip, now using temperature sensor or
The temperature of chip each several part in person's thermal sensation camera dynamic monitoring three-dimensional flash memory, then checks series of temperature measured value or direct
Analyze thermal sensation image, default significant data be assigned to the low block region of temperature, with ensure default significant data integrality and
Stability.
In step s 2, according to the default cooling of uniformity and operating system for monitoring chip each several part Temperature Distribution
Model selects the default method of salary distribution to carry out default significant data write operation.
In the present embodiment, the uniformity of the chip each several part Temperature Distribution is uniformly distributed non-with temperature including temperature
Be uniformly distributed, the cooling model includes strong cooling model and weak cooling model, therefore, according to uniformity of temperature profile whether and
Cool model it is whether powerful, build 4 kinds of different block allocation strategies, it is ensured that under different scenes be used for store physical block have compared with
High stability, prevents default significant data from producing mistake.
In the present embodiment, in view of chip is to the sensitiveness of temperature in three-dimensional flash memory, high temperature can destroy default important number
According to stability and integrality, so for some default significant datas, such as the metadata of file system, address of cache file
Deng, it is necessary to take some strategies to be assigned on the relatively low physical block of temperature, so that its reliability is guaranteed, wherein to default
In the allocation strategy of significant data, following two data structures have been related to：
(1) quaternary tree, is the quaternary tree based on 3D flash chip architectural characteristics, according to the architectural characteristic of 3D flash chips,
3D flash memories can be regarded as to a 3D cube, this cube includes α layers of layers, and every layer of layer includes β planes,
And each planes is formed by connecting by γ physical block (block), one layer of layer in selection flash chip is (such as
Layer_0), 2 × 2 dies are divided into, are then selected wherein the minimum die of temperature (such as die_2), continue to draw it
It is divided into 2 × 2 planes, the minimum plane (such as plane_1) of selection temperature, is divided into 2 × 2 blocks, and most
The minimum block regions of temperature are navigated to eventually, default significant data is write, and according to actual needs, can further be divided, after
The subtree of continuous generation more layers, the default important data structures are applied to the weaker situation of cooling model；
(2) queue, is the queue of the temperature proximity for physical block, the 3D flash chips of technology is captured using electric charge,
It is very sensitive to temperature, the proximity based on temperature, and adjacent physical block always has similar temperature, important to preset
Data perform write operation after, the temperature in the block region can rise, when write operation is very frequent, temperature also can rapid increase,
Every time during write-in, write-in block is distributed for it on different layer, the accumulation of heat can be avoided and abrasion equilibrium is improved, in order to
Ensure that the n write operation closed on is not repeatedly written same layer layer, queue structure can be used, it is assumed that have N layers of layer, team
Row Q is used to deposit this selected layer, and the layer will not be chosen in N-size (Q) sub-distribution backward, wherein
Size (Q) refers to the layer numbers included in queue Q, until all layer are written, the principle according still further to first in first out makes
With layer, reason for this is that in the case where cooling model is powerful, can be reserved to the upper one block region write enough
Time cooling, so that next time performs write-in to the block, the default important data structures are applied to the powerful situation of model that cools.
Referring to Fig. 2, monitoring that chip each several part temperature is uniformly distributed and presets for Fig. 1 in an embodiment of the present invention
Cooling model for strong cooling model when default significant data distribution method flow chart.
In the present embodiment, if monitoring chip each several part uniformity of temperature profile, illustrate chip each several part temperature compared with
It is close, therefore when for default significant data distribution physical block, the required precision that block is searched and positioned can be reduced, such as only
Just complete the positioning of write-in block using two layers of quaternary tree, and temperature distributing disproportionation it is even when complete block using three layers of quaternary tree
Positioning.In the present embodiment, if default cooling model is strong cooling model, then show that heating is slow, cooling is quick.
In step s 201, judge whether write operation is frequent；
If infrequently, comparing the temperature of each layer layer on chip in the three-dimensional flash memory in step S202, to obtain
The minimum layer low_temp_layer of temperature；
In step S203, the minimum layer low_temp_layer of temperature is divided half-and-half to obtain two subregion die；
In step S204, compare two subregion die temperature, and navigate to the minimum subregion low_temp_ of temperature
die；
In step S205, the minimum subregion low_temp_die of temperature is divided into n*n grid；
In step S206, m sample point is randomly choosed in n*n grid, and compares the temperature of m sample point, its
In, n*n/2<m<n*n；
In step S207, the minimum sample point of temperature is obtained, and write in the region where the minimum sample point of temperature
Enter default significant data.
When write operation is frequent, using the queue in above-mentioned default important data structures, write every time at different layer layers
Enter, it is to avoid the accumulation of heat and raising abrasion equilibrium, while time enough cooling is reserved in the block region to this write-in, specifically
, then judge whether the three-dimensional sudden strain of a muscle is equal to using the data queue size size (Q) of queue data structure in step S208
Deposit the number of plies layer_num of interior chip；
If being equal to, show that all layers of layer are at least written once in chip, now then take in step S209
Layer in dequeue positioned at head of the queue is as the minimum layer low_temp_layer of temperature, because it fully cools, then
Then perform step S203-S207, i.e., the minimum layer low_temp_layer of temperature is divided half-and-half to obtain two subregions
Die, compares two subregion die temperature, and navigates to the minimum subregion low_temp_die of temperature, by the subregion that temperature is minimum
Low_temp_die is divided into n*n grid, m sample point is randomly choosed in n*n grid, and compare m sample point
Temperature, obtains the minimum sample point of temperature, and writes default significant data in the region where the minimum sample point of temperature, its
In, n*n/2<m<n*n；
If being not equal to, in step S210, compare the temperature of each layer layer on chip in addition to queue Q to obtain temperature
The minimum layer low_temp_layer of degree.
In step S211, then the tail of the queue that the minimum layer low_temp_layer of temperature is inserted in queue transfers to hold
Row step S203-S207, i.e., divided half-and-half the minimum layer low_temp_layer of temperature to obtain two subregion die, than
Compared with two subregion die temperature, and the minimum subregion low_temp_die of temperature is navigated to, by the subregion low_ that temperature is minimum
Temp_die is divided into n*n grid, and m sample point is randomly choosed in n*n grid, and compares the temperature of m sample point,
The minimum sample point of temperature is obtained, and default significant data is write in the region where the minimum sample point of temperature, wherein, n*
n/2<m<n*n。
Referring to Fig. 3, monitoring that chip each several part temperature is uniformly distributed and presets for Fig. 1 in an embodiment of the present invention
Cooling model for it is weak cooling model when default significant data distribution method flow chart.
In the present embodiment, if default cooling model is weak cooling model, then show that heating is quick, cooling is slow
Slowly, now using two layers of quad-tree partition layer layer and subregion die.
In step S301, temperature in use sensor or thermal sensation camera thermometric compare each on the three-dimensional flash memory chip
Layer layer temperature.
In step s 302, the minimum layer low_temp_layer of temperature is obtained, and by the minimum layer low_temp_ of temperature
Layer is divided into n*n subregion die, wherein, due to being quaternary tree, n now takes 2.
In step S303, compare each subregion die temperature, and navigate to the minimum subregion low_temp_ of temperature
die。
In step s 304, the minimum subregion low_temp_die bandings of temperature are divided into m sub-district.
In step S305, a sample point is respectively taken using each sub-district of equidistant sampling method in m sub-district, with shape
Into m sample point.In the present embodiment, using equidistant sampling method the reason for is uniformity of temperature profile, and contiguous physical block
Temperature there is proximity, a sample point can be taken at a certain distance, goes to represent the segment distance with the temperature of the sample point
The temperature of all physical blocks.
In step S306, p sample point is taken to be compared temperature at random in m sample point, wherein, m/2<p<m.
In step S307, the minimum sample point of temperature is obtained, and write in the region where the minimum sample point of temperature
Enter default significant data.
Referring to Fig. 4, monitoring chip each several part temperature non-uniform Distribution and pre- for Fig. 1 in an embodiment of the present invention
If cooling model for strong cooling model when default significant data distribution method flow chart.
In the present embodiment, temperature non-uniform Distribution shows that chip each several part temperature difference is larger, default cooling mould
Type then shows that heating is slow for strong cooling model, and cooling is quick.
In step S401, judge whether write operation is frequent；
If infrequently, in step S402, each layer layer temperature on relatively more described three-dimensional flash memory chip；
In step S403, the minimum layer low_temp_layer of temperature is obtained, and by the minimum layer low_temp_ of temperature
Layer is divided into n*n subregion die；
In step s 404, compare each subregion die temperature, and navigate to the minimum subregion low_temp_ of temperature
die；
In step S405, the minimum subregion low_temp_die of temperature is divided into m*m grid；
In step S406, p sample point is randomly selected in m*m grid and is compared temperature, wherein, m*m/2<p<
m*m；
In step S 407, the minimum sample point of temperature is obtained, and is write in the region where the minimum sample point of temperature
Enter default significant data.
When write operation is frequent, equally using the queue in above-mentioned data structure, write every time at different layer layers,
Avoid the accumulation of heat and improve abrasion equilibrium, while time enough cooling is reserved in the block region to this write-in, specifically,
Then in step S408, judge whether the three-dimensional flash memory is equal to using data queue size (Q) size of queue data structure
The number of plies layer_num of interior chip；
If being equal to, in step S409, take out the layer in queue positioned at head of the queue and be used as the minimum layer low_temp_ of temperature
Layer, because it fully cools, then performs step S403-S407, i.e., by the minimum layer low_temp_ of temperature
Layer is divided into n*n subregion die, compares each subregion die temperature, and navigates to the minimum subregion low_temp_ of temperature
Die, the minimum subregion low_temp_die of temperature is divided into m*m grid, p sample is randomly selected in m*m grid
Point carries out C.T, obtains the minimum sample point of temperature, and write-in is default in the region where the minimum sample point of temperature
Significant data, wherein, m*m/2<p<m*m；
If being not equal to, in step S410, compare the temperature of each layer layer on chip in addition to queue Q to obtain temperature
The minimum layer low_temp_layer of degree.
In step S411, then the tail of the queue that the minimum layer low_temp_layer of temperature is inserted in queue performs step
Rapid S403-S407, i.e., be divided into n*n subregion die by the minimum layer low_temp_layer of temperature, compare each subregion die
Temperature, and navigate to the minimum subregion low_temp_die of temperature, the minimum subregion low_temp_die of temperature be divided into
M*m grid, p sample point is randomly selected in m*m grid and is compared temperature, obtains the minimum sample point of temperature, and
Default significant data is write in the region where the minimum sample point of temperature, wherein, m*m/2<p<m*m.
Referring to Fig. 5, monitoring chip each several part temperature non-uniform Distribution and pre- for Fig. 1 in an embodiment of the present invention
If cooling model for it is weak cooling model when default significant data distribution method flow chart.
In the present embodiment, temperature non-uniform Distribution shows that chip each several part temperature difference is larger, if default drop
Temperature model is weak cooling model, then shows that heating is quick, cooling is slow, now using two layers quad-tree partition layer layer and
Subregion die.
In step S501, temperature in use sensor or thermal sensation camera thermometric compare each on the three-dimensional flash memory chip
Layer layer temperature；
In step S502, the minimum layer low_temp_layer of temperature is obtained, and by the minimum layer low_temp_ of temperature
Layer is divided into n*n subregion die；
In step S503, compare each subregion die temperature, and navigate to the minimum subregion low_temp_ of temperature
die；
In step S504, the minimum subregion low_temp_die of temperature is divided into n*n sub-district plane；
In step S505, compare each sub-district plane temperature, and navigate to the minimum sub-district low_temp_ of temperature
plane；
In step S506, m sample is taken in the minimum sub-district low_temp_plane of temperature using accidental sampling
Point；
In step s 507, the temperature of relatively more all m sample points；
In step S508, the minimum sample point of temperature is obtained, and write in the region where the minimum sample point of temperature
Enter default significant data.In the present embodiment, write after default significant data, be last_access by the block zone marker,
Represent that the block region is accessed recently, first directly ignore it during next selection block, and in selection next time, use last_
Access marks next selected block region, cancels the mark in former selection block region, because a certain piece is chosen row write of going forward side by side
After entering, temperature can rise, and due to the proximity of temperature, influence whether a whole region up and down for the block, in cooling model
In the case of weaker, not repeating to write same region can be to the time for being written into the reserved cooling of block.
A kind of significant data distribution method based on temperature that the present invention is provided, temperature in use sensor or thermal sensation are mutually motor-driven
State monitors the temperature of 3D flash chips, and default significant data is assigned into the low block region of temperature, to ensure default significant data
Integrality and stability.The technical scheme that the present invention is provided according to uniformity of temperature profile whether and cooling model it is powerful with
It is no, build 4 kinds of different block allocation strategies, it is ensured that the physical block for being used to store under different scenes has higher stability, prevents
Only preset significant data and produce mistake, while aid in realizing the Search and Orientation of block using quaternary tree and queue data structure, can
To navigate to the block region that temperature is low, and the situation for performing write operation repeatedly in same area is avoided, reduce the product of heat
It is tired, abrasion equilibrium is improved, the life-span of chip is also extended.
Referring to Fig. 6, showing in an embodiment of the present invention the structure of the significant data distribution system 10 based on temperature
Schematic diagram.
In the present embodiment, the significant data distribution system 10 based on temperature, it is main to include surveying applied to three-dimensional flash memory
Warm module 11, distribute module 12.Wherein, distribute module 12 specifically includes the first distribution sub module 121, the second distribution sub module
122nd, the 3rd distribution sub module 123 and the 4th distribution sub module 124.
Temperature measurement module 11, for when the operating system of the three-dimensional flash memory sends the request for writing default significant data, supervising
Survey the temperature of chip each several part in the three-dimensional flash memory；
Distribute module 12, for default according to the uniformity and operating system that monitor chip each several part Temperature Distribution
The model that cools selects the default method of salary distribution to carry out default significant data write operation.
In the present embodiment, the temperature of temperature sensor or thermal sensation camera dynamic monitoring chip each several part, institute are utilized
The uniformity for stating chip each several part Temperature Distribution is uniformly distributed including temperature and included with temperature non-uniform Distribution, the cooling model
Strong cooling model and weak cooling model.
In the present embodiment, monitoring that chip each several part temperature is uniformly distributed and default cooling model is strong cooling
During model, first distribution sub module 121 is used for：
Judge whether write operation is frequent；
If infrequently, in the three-dimensional flash memory on chip each layer temperature, to obtain the layer that temperature is minimum；
The minimum layer of temperature is divided half-and-half to obtain two subregions；
Compare the temperature of two subregions, and navigate to the minimum subregion of temperature；
The minimum subregion of temperature is divided into n*n grid；
M sample point is randomly choosed in n*n grid, and compares the temperature of m sample point；
The minimum sample point of temperature is obtained, and important number is preset in write-in in the region where the minimum sample point of temperature
According to.
In the present embodiment, first distribution sub module 121 is additionally operable to：
When write operation is frequent, judge whether the three-dimensional sudden strain of a muscle is equal to using the data queue size of queue data structure
Deposit the number of plies of interior chip；
If being equal to, the layer for being located at head of the queue in queue is taken out as the minimum layer of temperature, by the layer progress pair that temperature is minimum
Half point compares the temperature of two subregions to obtain two subregions, and navigates to the minimum subregion of temperature, by the subregion that temperature is minimum
N*n grid is divided into, m sample point is randomly choosed in n*n grid, and compares the temperature of m sample point, temperature is obtained
Minimum sample point, and the default significant data of write-in in the region where the minimum sample point of temperature；
If being not equal to, compare the temperature of each layer on chip in addition to queue to obtain the layer that temperature is minimum, by temperature most
Tail of the queue in low layer insertion queue, and the minimum layer of temperature is divided half-and-half to obtain two subregions, compare two subregions
Temperature, and navigate to the minimum subregion of temperature, the minimum subregion of temperature be divided into n*n grid, in n*n grid with
Machine selects m sample point, and compares the temperature of m sample point, obtains the minimum sample point of temperature, and in the minimum sample of temperature
The default significant data of write-in in region where point.
In the present embodiment, monitoring that chip each several part temperature is uniformly distributed and default cooling model is weak cooling
During model, second distribution sub module 122 is used for：
The temperature of each layer on temperature in use sensor or thermal sensation camera thermometric, relatively more described three-dimensional flash memory chip；
The minimum layer of temperature is obtained, and the minimum layer of temperature is divided into n*n subregion；
Compare the temperature of each subregion, and navigate to the minimum subregion of temperature；
The minimum subregion banding of temperature is divided into m sub-district；
A sample point is respectively taken using each sub-district of equidistant sampling method in m sub-district, to form m sample point；
P sample point is taken to be compared temperature at random in m sample point；
The minimum sample point of temperature is obtained, and important number is preset in write-in in the region where the minimum sample point of temperature
According to.
In the present embodiment, monitoring that chip each several part temperature non-uniform Distribution and default cooling model are strong drop
During temperature model, the 3rd distribution sub module 123 is used for：
Judge whether write operation is frequent；
If infrequently, on the three-dimensional flash memory chip each layer temperature；
The minimum layer of temperature is obtained, and the minimum layer of temperature is divided into n*n subregion；
Compare the temperature of each subregion, and navigate to the minimum subregion of temperature；
The minimum subregion of temperature is divided into m*m grid；
P sample point is randomly selected in m*m grid and is compared temperature；
The minimum sample point of temperature is obtained, and important number is preset in write-in in the region where the minimum sample point of temperature
According to.
In the present embodiment, the 3rd distribution sub module 123 is additionally operable to：
When write operation is frequent, judge whether the three-dimensional sudden strain of a muscle is equal to using the data queue size of queue data structure
Deposit the number of plies of interior chip；
If being equal to, the layer for being located at head of the queue in queue is taken out as the minimum layer of temperature, the minimum layer of temperature is divided into
N*n subregion, compares the temperature of each subregion, and navigates to the minimum subregion of temperature, and the minimum subregion of temperature is divided into m*m
Grid, p sample point is randomly selected in m*m grid and is compared temperature, the minimum sample point of temperature is obtained, and in temperature
The default significant data of write-in in region where the minimum sample point of degree；
If being not equal to, compare the temperature of each layer on chip in addition to queue to obtain the layer that temperature is minimum, by temperature most
Tail of the queue in low layer insertion queue, and the minimum layer of temperature is divided into n*n subregion, compare the temperature of each subregion, and
The minimum subregion of temperature is navigated to, the minimum subregion of temperature is divided into m*m grid, p are randomly selected in m*m grid
Sample point is compared temperature, obtains the minimum sample point of temperature, and write in the region where the minimum sample point of temperature
Default significant data.
In the present embodiment, monitoring that chip each several part temperature non-uniform Distribution and default cooling model are weak drop
During temperature model, the 4th distribution sub module 124 is used for：
The temperature of each layer on temperature in use sensor or thermal sensation camera thermometric, relatively more described three-dimensional flash memory chip；
The minimum layer of temperature is obtained, and the minimum layer of temperature is divided into n*n subregion；
Compare the temperature of each subregion, and navigate to the minimum subregion of temperature；
The minimum subregion of temperature is divided into n*n sub-district；
Compare the temperature of each sub-district, and navigate to the minimum sub-district of temperature；
Using accidental sampling m sample point is taken in the minimum sub-district of temperature；
Compare the temperature of all m sample points；
The minimum sample point of temperature is obtained, and important number is preset in write-in in the region where the minimum sample point of temperature
According to.
A kind of significant data distribution system 10 based on temperature that the present invention is provided, temperature in use sensor or thermal sensation camera
The temperature of dynamic monitoring 3D flash chips, the low block region of temperature is assigned to by default significant data, to ensure to preset important number
According to integrality and stability.The technical scheme that the present invention is provided according to uniformity of temperature profile whether and cooling model it is powerful with
It is no, build 4 kinds of different block allocation strategies, it is ensured that the physical block for being used to store under different scenes has higher stability, prevents
Only preset significant data and produce mistake, while aid in realizing the Search and Orientation of block using quaternary tree and queue data structure, can
To navigate to the block region that temperature is low, and the situation for performing write operation repeatedly in same area is avoided, reduce the product of heat
It is tired, abrasion equilibrium is improved, the life-span of chip is also extended.
It is worth noting that, in above-described embodiment, included unit is simply divided according to function logic,
But above-mentioned division is not limited to, as long as corresponding function can be realized；In addition, the specific name of each functional unit
Only to facilitate mutually distinguishing, the protection domain being not intended to limit the invention.
In addition, one of ordinary skill in the art will appreciate that realizing all or part of step in the various embodiments described above method
It can be by program to instruct the hardware of correlation to complete, corresponding program can be stored in embodied on computer readable storage Jie
In matter, described storage medium, such as ROM/RAM, disk or CD.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
Any modifications, equivalent substitutions and improvements made within refreshing and principle etc., should be included in the scope of the protection.
Claims (10)
1. a kind of significant data distribution method based on temperature, applied to three-dimensional flash memory, it is characterised in that methods described includes：
When the operating system of the three-dimensional flash memory sends the request for writing default significant data, chip in the three-dimensional flash memory is monitored
The temperature of each several part；
It is default according to the default cooling model selection of the uniformity and operating system that monitor chip each several part Temperature Distribution
The method of salary distribution carries out the default significant data write operation.
2. the significant data distribution method based on temperature as claimed in claim 1, it is characterised in that using temperature sensor or
The temperature of person's thermal sensation camera dynamic monitoring chip each several part, the uniformity of the chip each several part Temperature Distribution is uniform including temperature
Distribution and temperature non-uniform Distribution, the cooling model include strong cooling model and weak cooling model；
When monitoring that chip each several part temperature is uniformly distributed and default cooling model is strong cooling model, the selection is preset
The method of salary distribution carry out the default significant data write operation the step of specifically include：
Judge whether write operation is frequent；
If infrequently, in the three-dimensional flash memory on chip each layer temperature, to obtain the layer that temperature is minimum；
The minimum layer of temperature is divided half-and-half to obtain two subregions；
Compare the temperature of two subregions, and navigate to the minimum subregion of temperature；
The minimum subregion of temperature is divided into n*n grid；
M sample point is randomly choosed in n*n grid, and compares the temperature of m sample point；
The minimum sample point of temperature is obtained, and writes in the region where the minimum sample point of temperature and described to preset important number
According to；
The step of default method of salary distribution of selection carries out the default significant data write operation specifically also includes：
When write operation is frequent, judge whether be equal in the three-dimensional flash memory using the data queue size of queue data structure
The number of plies of chip；
If being equal to, the layer for being located at head of the queue in queue is taken out as the minimum layer of temperature, the minimum layer of temperature is divided half-and-half
To obtain two subregions, compare the temperature of two subregions, and navigate to the minimum subregion of temperature, the minimum subregion of temperature is divided
For n*n grid, m sample point is randomly choosed in n*n grid, and compares the temperature of m sample point, temperature is obtained minimum
Sample point, and write in the region where the minimum sample point of temperature the default significant data；
If being not equal to, compare the temperature of each layer on chip in addition to queue to obtain the layer that temperature is minimum, temperature is minimum
Tail of the queue in layer insertion queue, and the minimum layer of temperature is divided half-and-half to obtain two subregions, compare the temperature of two subregions
Degree, and the minimum subregion of temperature is navigated to, the minimum subregion of temperature is divided into n*n grid, selected at random in n*n grid
M sample point is selected, and compares the temperature of m sample point, the minimum sample point of temperature is obtained, and in the minimum sample point institute of temperature
Region in write the default significant data.
3. the significant data distribution method based on temperature as claimed in claim 1, it is characterised in that using temperature sensor or
The temperature of person's thermal sensation camera dynamic monitoring chip each several part, the uniformity of the chip each several part Temperature Distribution is uniform including temperature
Distribution and temperature non-uniform Distribution, the cooling model include strong cooling model and weak cooling model；
When monitoring that chip each several part temperature is uniformly distributed and default cooling model is weak cooling model, the selection is preset
The method of salary distribution carry out the default significant data write operation the step of specifically include：
The temperature of each layer on temperature in use sensor or thermal sensation camera thermometric, relatively more described three-dimensional flash memory chip；
The minimum layer of temperature is obtained, and the minimum layer of temperature is divided into n*n subregion；
Compare the temperature of each subregion, and navigate to the minimum subregion of temperature；
The minimum subregion banding of temperature is divided into m sub-district；
A sample point is respectively taken using each sub-district of equidistant sampling method in m sub-district, to form m sample point；
P sample point is taken to be compared temperature at random in m sample point；
The minimum sample point of temperature is obtained, and writes in the region where the minimum sample point of temperature and described to preset important number
According to.
4. the significant data distribution method based on temperature as claimed in claim 1, it is characterised in that using temperature sensor or
The temperature of person's thermal sensation camera dynamic monitoring chip each several part, the uniformity of the chip each several part Temperature Distribution is uniform including temperature
Distribution and temperature non-uniform Distribution, the cooling model include strong cooling model and weak cooling model；
When monitoring chip each several part temperature non-uniform Distribution and default cooling model is strong cooling model, the selection is pre-
If the method for salary distribution carry out the default significant data write operation the step of specifically include：
Judge whether write operation is frequent；
If infrequently, on the three-dimensional flash memory chip each layer temperature；
The minimum layer of temperature is obtained, and the minimum layer of temperature is divided into n*n subregion；
Compare the temperature of each subregion, and navigate to the minimum subregion of temperature；
The minimum subregion of temperature is divided into m*m grid；
P sample point is randomly selected in m*m grid and is compared temperature；
The minimum sample point of temperature is obtained, and writes in the region where the minimum sample point of temperature and described to preset important number
According to；
The step of default method of salary distribution of selection carries out the default significant data write operation specifically includes：
When write operation is frequent, judge whether be equal in the three-dimensional flash memory using the data queue size of queue data structure
The number of plies of chip；
If being equal to, the layer for being located at head of the queue in queue is taken out as the minimum layer of temperature, the minimum layer of temperature is divided into n*n
Subregion, compares the temperature of each subregion, and navigates to the minimum subregion of temperature, and the minimum subregion of temperature is divided into m*m net
Lattice, p sample point is randomly selected in m*m grid and is compared temperature, obtains the minimum sample point of temperature, and in temperature most
The default significant data is write in region where low sample point；
If being not equal to, compare the temperature of each layer on chip in addition to queue to obtain the layer that temperature is minimum, temperature is minimum
Tail of the queue in layer insertion queue, and the minimum layer of temperature is divided into n*n subregion, compare the temperature of each subregion, and position
To the subregion that temperature is minimum, the minimum subregion of temperature is divided into m*m grid, p sample is randomly selected in m*m grid
Point carries out C.T, obtains the minimum sample point of temperature, and write-in is described in the region where the minimum sample point of temperature
Default significant data.
5. the significant data distribution method based on temperature as claimed in claim 1, it is characterised in that using temperature sensor or
The temperature of person's thermal sensation camera dynamic monitoring chip each several part, the uniformity of the chip each several part Temperature Distribution is uniform including temperature
Distribution and temperature non-uniform Distribution, the cooling model include strong cooling model and weak cooling model；
When monitoring chip each several part temperature non-uniform Distribution and default cooling model is weak cooling model, the selection is pre-
If the method for salary distribution carry out the default significant data write operation the step of specifically include：
The temperature of each layer on temperature in use sensor or thermal sensation camera thermometric, relatively more described three-dimensional flash memory chip；
The minimum layer of temperature is obtained, and the minimum layer of temperature is divided into n*n subregion；
Compare the temperature of each subregion, and navigate to the minimum subregion of temperature；
The minimum subregion of temperature is divided into n*n sub-district；
Compare the temperature of each sub-district, and navigate to the minimum sub-district of temperature；
Using accidental sampling m sample point is taken in the minimum sub-district of temperature；
Compare the temperature of all m sample points；
The minimum sample point of temperature is obtained, and writes in the region where the minimum sample point of temperature and described to preset important number
According to.
6. a kind of significant data distribution system based on temperature, applied to three-dimensional flash memory, it is characterised in that the system includes：
Temperature measurement module, for when the operating system of the three-dimensional flash memory sends the request for writing default significant data, monitoring to be described
The temperature of chip each several part in three-dimensional flash memory；
Distribute module, for according to the default cooling mould of uniformity and operating system for monitoring chip each several part Temperature Distribution
Type selects the default method of salary distribution to carry out the default significant data write operation.
7. the significant data distribution system based on temperature as claimed in claim 6, it is characterised in that using temperature sensor or
The temperature of person's thermal sensation camera dynamic monitoring chip each several part, the uniformity of the chip each several part Temperature Distribution is uniform including temperature
Distribution and temperature non-uniform Distribution, the cooling model include strong cooling model and weak cooling model；
When monitoring that chip each several part temperature is uniformly distributed and default cooling model is strong cooling model, the distribute module
The first distribution sub module is specifically included, first distribution sub module is used for：
Judge whether write operation is frequent；
If infrequently, in the three-dimensional flash memory on chip each layer temperature, to obtain the layer that temperature is minimum；
The minimum layer of temperature is divided half-and-half to obtain two subregions；
Compare the temperature of two subregions, and navigate to the minimum subregion of temperature；
The minimum subregion of temperature is divided into n*n grid；
M sample point is randomly choosed in n*n grid, and compares the temperature of m sample point；
The minimum sample point of temperature is obtained, and writes in the region where the minimum sample point of temperature and described to preset important number
According to；
First distribution sub module is additionally operable to：
When write operation is frequent, judge whether be equal in the three-dimensional flash memory using the data queue size of queue data structure
The number of plies of chip；
If being equal to, the layer for being located at head of the queue in queue is taken out as the minimum layer of temperature, the minimum layer of temperature is divided half-and-half
To obtain two subregions, compare the temperature of two subregions, and navigate to the minimum subregion of temperature, the minimum subregion of temperature is divided
For n*n grid, m sample point is randomly choosed in n*n grid, and compares the temperature of m sample point, temperature is obtained minimum
Sample point, and write in the region where the minimum sample point of temperature the default significant data；
If being not equal to, compare the temperature of each layer on chip in addition to queue to obtain the layer that temperature is minimum, temperature is minimum
Tail of the queue in layer insertion queue, and the minimum layer of temperature is divided half-and-half to obtain two subregions, compare the temperature of two subregions
Degree, and the minimum subregion of temperature is navigated to, the minimum subregion of temperature is divided into n*n grid, selected at random in n*n grid
M sample point is selected, and compares the temperature of m sample point, the minimum sample point of temperature is obtained, and in the minimum sample point institute of temperature
Region in write the default significant data.
8. the significant data distribution system based on temperature as claimed in claim 6, it is characterised in that using temperature sensor or
The temperature of person's thermal sensation camera dynamic monitoring chip each several part, the uniformity of the chip each several part Temperature Distribution is uniform including temperature
Distribution and temperature non-uniform Distribution, the cooling model include strong cooling model and weak cooling model；
When monitoring that chip each several part temperature is uniformly distributed and default cooling model is weak cooling model, the distribute module
The second distribution sub module is specifically included, second distribution sub module is used for：
The temperature of each layer on temperature in use sensor or thermal sensation camera thermometric, relatively more described three-dimensional flash memory chip；
The minimum layer of temperature is obtained, and the minimum layer of temperature is divided into n*n subregion；
Compare the temperature of each subregion, and navigate to the minimum subregion of temperature；
The minimum subregion banding of temperature is divided into m sub-district；
A sample point is respectively taken using each sub-district of equidistant sampling method in m sub-district, to form m sample point；
P sample point is taken to be compared temperature at random in m sample point；
The minimum sample point of temperature is obtained, and writes in the region where the minimum sample point of temperature and described to preset important number
According to.
9. the significant data distribution system based on temperature as claimed in claim 6, it is characterised in that using temperature sensor or
The temperature of person's thermal sensation camera dynamic monitoring chip each several part, the uniformity of the chip each several part Temperature Distribution is uniform including temperature
Distribution and temperature non-uniform Distribution, the cooling model include strong cooling model and weak cooling model；
When monitoring chip each several part temperature non-uniform Distribution and default cooling model is strong cooling model, the distribution mould
Block specifically includes the 3rd distribution sub module, and the 3rd distribution sub module is used for：
Judge whether write operation is frequent；
If infrequently, on the three-dimensional flash memory chip each layer temperature；
The minimum layer of temperature is obtained, and the minimum layer of temperature is divided into n*n subregion；
Compare the temperature of each subregion, and navigate to the minimum subregion of temperature；
The minimum subregion of temperature is divided into m*m grid；
P sample point is randomly selected in m*m grid and is compared temperature；
The minimum sample point of temperature is obtained, and writes in the region where the minimum sample point of temperature and described to preset important number
According to；
3rd distribution sub module is additionally operable to：
When write operation is frequent, judge whether be equal in the three-dimensional flash memory using the data queue size of queue data structure
The number of plies of chip；
If being equal to, the layer for being located at head of the queue in queue is taken out as the minimum layer of temperature, the minimum layer of temperature is divided into n*n
Subregion, compares the temperature of each subregion, and navigates to the minimum subregion of temperature, and the minimum subregion of temperature is divided into m*m net
Lattice, p sample point is randomly selected in m*m grid and is compared temperature, obtains the minimum sample point of temperature, and in temperature most
The default significant data is write in region where low sample point；
If being not equal to, compare the temperature of each layer on chip in addition to queue to obtain the layer that temperature is minimum, temperature is minimum
Tail of the queue in layer insertion queue, and the minimum layer of temperature is divided into n*n subregion, compare the temperature of each subregion, and position
To the subregion that temperature is minimum, the minimum subregion of temperature is divided into m*m grid, p sample is randomly selected in m*m grid
Point carries out C.T, obtains the minimum sample point of temperature, and write-in is described in the region where the minimum sample point of temperature
Default significant data.
10. the significant data distribution system as claimed in claim 6 based on temperature, it is characterised in that utilize temperature sensor
Or the temperature of thermal sensation camera dynamic monitoring chip each several part, it is equal that the uniformity of the chip each several part Temperature Distribution includes temperature
Even distribution and temperature non-uniform Distribution, the cooling model include strong cooling model and weak cooling model；
When monitoring chip each several part temperature non-uniform Distribution and default cooling model is weak cooling model, the distribution mould
Block specifically includes the 4th distribution sub module, and the 4th distribution sub module is used for：
The temperature of each layer on temperature in use sensor or thermal sensation camera thermometric, relatively more described three-dimensional flash memory chip；
The minimum layer of temperature is obtained, and the minimum layer of temperature is divided into n*n subregion；
Compare the temperature of each subregion, and navigate to the minimum subregion of temperature；
The minimum subregion of temperature is divided into n*n sub-district；
Compare the temperature of each sub-district, and navigate to the minimum sub-district of temperature；
Using accidental sampling m sample point is taken in the minimum sub-district of temperature；
Compare the temperature of all m sample points；
The minimum sample point of temperature is obtained, and writes in the region where the minimum sample point of temperature and described to preset important number
According to.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710316728.9A CN107239409B (en) | 2017-05-08 | 2017-05-08 | Temperature-based important data distribution method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710316728.9A CN107239409B (en) | 2017-05-08 | 2017-05-08 | Temperature-based important data distribution method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107239409A true CN107239409A (en) | 2017-10-10 |
| CN107239409B CN107239409B (en) | 2020-12-29 |
Family
ID=59984988
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710316728.9A Active CN107239409B (en) | 2017-05-08 | 2017-05-08 | Temperature-based important data distribution method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107239409B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109634960A (en) * | 2018-12-04 | 2019-04-16 | 深圳大学 | Key assignments date storage method, device, equipment and storage medium |
| CN110531935A (en) * | 2019-08-29 | 2019-12-03 | 北京计算机技术及应用研究所 | A kind of SLC Flash abrasion equilibrium method based on Physical Page relative temperature |
| CN110874184A (en) * | 2018-09-03 | 2020-03-10 | 合肥沛睿微电子股份有限公司 | Flash memory controller and related electronic device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090171513A1 (en) * | 2007-12-27 | 2009-07-02 | Kabushiki Kaisha Toshiba | Information Processing Apparatus and Semiconductor Storage Drive |
| CN102298966A (en) * | 2010-05-31 | 2011-12-28 | 三星电子株式会社 | Nonvolatile memory device, system and programming method with dynamic verification mode selection |
| CN103176799A (en) * | 2013-02-28 | 2013-06-26 | 山东大学 | Temperature-sensitive mixed storage framework and data distribution strategy thereof |
| CN103890850A (en) * | 2011-09-30 | 2014-06-25 | 英特尔公司 | Dynamic operations for 3D stacked memory using thermal data |
| CN105426316A (en) * | 2015-11-09 | 2016-03-23 | 北京大学 | Quota control temperature based racetrack memory chip and control method therefor |
| CN105677578A (en) * | 2016-01-08 | 2016-06-15 | 深圳大学 | Control method and system for 3D flash memory |
| US20160239235A1 (en) * | 2015-02-17 | 2016-08-18 | Woonjae Chung | Storage devices, memory systems and operating methods thereof |
-
2017
- 2017-05-08 CN CN201710316728.9A patent/CN107239409B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090171513A1 (en) * | 2007-12-27 | 2009-07-02 | Kabushiki Kaisha Toshiba | Information Processing Apparatus and Semiconductor Storage Drive |
| CN102298966A (en) * | 2010-05-31 | 2011-12-28 | 三星电子株式会社 | Nonvolatile memory device, system and programming method with dynamic verification mode selection |
| CN103890850A (en) * | 2011-09-30 | 2014-06-25 | 英特尔公司 | Dynamic operations for 3D stacked memory using thermal data |
| CN103176799A (en) * | 2013-02-28 | 2013-06-26 | 山东大学 | Temperature-sensitive mixed storage framework and data distribution strategy thereof |
| US20160239235A1 (en) * | 2015-02-17 | 2016-08-18 | Woonjae Chung | Storage devices, memory systems and operating methods thereof |
| CN105426316A (en) * | 2015-11-09 | 2016-03-23 | 北京大学 | Quota control temperature based racetrack memory chip and control method therefor |
| CN105677578A (en) * | 2016-01-08 | 2016-06-15 | 深圳大学 | Control method and system for 3D flash memory |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110874184A (en) * | 2018-09-03 | 2020-03-10 | 合肥沛睿微电子股份有限公司 | Flash memory controller and related electronic device |
| CN110874184B (en) * | 2018-09-03 | 2023-08-22 | 合肥沛睿微电子股份有限公司 | Flash memory controller and related electronic device |
| CN109634960A (en) * | 2018-12-04 | 2019-04-16 | 深圳大学 | Key assignments date storage method, device, equipment and storage medium |
| CN109634960B (en) * | 2018-12-04 | 2021-01-26 | 深圳大学 | Key value data storage method, device, equipment and storage medium |
| CN110531935A (en) * | 2019-08-29 | 2019-12-03 | 北京计算机技术及应用研究所 | A kind of SLC Flash abrasion equilibrium method based on Physical Page relative temperature |
| CN110531935B (en) * | 2019-08-29 | 2023-02-03 | 北京计算机技术及应用研究所 | SLC flash memory wear leveling method based on physical page relative temperature |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107239409B (en) | 2020-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102597640B1 (en) | Optimized hopscotch multiple hash tables for efficient memory in-line deduplication application | |
| US20210312952A1 (en) | Interconnect architecture for three-dimensional processing systems | |
| US11556250B2 (en) | Method and apparatus for temperature-gradient aware data-placement for 3D stacked DRAMs | |
| CN107239409A (en) | A kind of significant data distribution method and its system based on temperature | |
| Pathak et al. | Through-silicon-via management during 3D physical design: When to add and how many? | |
| KR102597630B1 (en) | Virtual bucket multiple hash tables for efficient memory in-line deduplication application | |
| JP2017188096A (en) | Deduplication memory module and memory deduplication method therefor | |
| US9947386B2 (en) | Thermal aware data placement and compute dispatch in a memory system | |
| CN103828498A (en) | System and method for sequential placement of cooling resources within data center layouts | |
| Zhu et al. | Integrated thermal analysis for processing in die-stacking memory | |
| TWI629642B (en) | Simulation system and method thereof and computing system including the simulation system | |
| US10482033B2 (en) | Method and device for controlling memory | |
| TW201741883A (en) | Method and dedupe module for enabling larger memory capacity than physical memory size | |
| US20150134877A1 (en) | Data storage system with passive partitioning in a secondary memory | |
| Perumal et al. | Phase-field study on the formation of first-neighbour topological clusters during the isotropic grain growth | |
| CN107580698A (en) | System and method for the complicating factors of the scheduling size that determines parallel processor kernel | |
| CN109460406A (en) | Data processing method and device | |
| EP3051448A1 (en) | Methods for distributing power in layout of ic | |
| CN106980540B (en) | Distributed multi-dimensional discrete data calculation method | |
| WO2016191964A1 (en) | Management method and device of file system | |
| CN102124452A (en) | Storage region management method, storage region allocation method and program | |
| US8826214B2 (en) | Implementing Z directional macro port assignment | |
| CN115510650A (en) | Temperature distribution data determination method, temperature distribution data determination device, computer equipment and storage medium | |
| CN107168894B (en) | Memory sharing access method and electronic equipment | |
| Chen et al. | A cluster-based reliability-and thermal-aware 3D floorplanning using redundant STSVs |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |