CN109635315B - Reliability modeling and design guiding method oriented to embryo hardware cell reuse strategy - Google Patents
Reliability modeling and design guiding method oriented to embryo hardware cell reuse strategy Download PDFInfo
- Publication number
- CN109635315B CN109635315B CN201811269642.6A CN201811269642A CN109635315B CN 109635315 B CN109635315 B CN 109635315B CN 201811269642 A CN201811269642 A CN 201811269642A CN 109635315 B CN109635315 B CN 109635315B
- Authority
- CN
- China
- Prior art keywords
- cell
- reliability
- array
- cells
- embryo
- 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.)
- Active
Links
- 238000013461 design Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 49
- 210000001161 mammalian embryo Anatomy 0.000 title claims abstract description 37
- 238000004458 analytical method Methods 0.000 claims abstract description 26
- 238000005457 optimization Methods 0.000 claims abstract description 22
- 238000004364 calculation method Methods 0.000 claims abstract description 8
- 230000008439 repair process Effects 0.000 claims description 44
- 230000006872 improvement Effects 0.000 claims description 11
- 238000013507 mapping Methods 0.000 claims description 2
- 238000013139 quantization Methods 0.000 claims description 2
- 230000001052 transient effect Effects 0.000 abstract description 38
- 230000001965 increasing effect Effects 0.000 abstract description 11
- 230000006870 function Effects 0.000 description 15
- 230000008569 process Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 239000011664 nicotinic acid Substances 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000004377 microelectronic Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012942 design verification Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 230000000191 radiation effect Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/39—Circuit design at the physical level
- G06F30/398—Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM]
Abstract
The invention discloses a reliability modeling and design guiding method for an embryo hardware cell reuse strategy, which comprises a reliability modeling method for the embryo hardware cell reuse strategy and a guiding method for optimizing the layout design of an embryo cell array under different space environments based on reliability model analysis. The invention takes the reusable fault cells as redundant cells to carry out reliability modeling, and solves the problem that the traditional reliability model can not analyze the influence of transient fault cells; the reliability model is built according to the k-out-of-m model, the sum of the redundant idle cell number of the array and the repairable times increased due to repairable transient faults is represented by the total number of equivalent idle cells in an equation, and the cell reuse strategy can be analyzed more accurately; aiming at fault characteristics under different space environments, transient fault proportion parameters are provided for calculation, reliability comparison under different fault characteristics is realized, and a guiding standard is provided for optimization selection.
Description
Technical Field
The invention relates to a reliability modeling and design guidance method for an embryo hardware cell reuse strategy, and belongs to the technical field of bionic self-repairing hardware design.
Background
With the rapid development of microelectronic technology, the integration level of semiconductor circuits is continuously improved, and the feature size of space electronic devices is smaller and smaller, so that the single event effect is easier to be caused. The single event effect, which is a radiation effect that causes abnormal changes in the state of the device when a single energetic particle passes through the sensitive region of a microelectronic device, has now become a major factor affecting the service life and performance of aerospace electronic systems. High energy particles in the space-borne radiation bombard memories or flip-flops, etc., causing the logic values of the digital circuits to flip, referred to as single event upset (Single Event Upset, SEU). Single event upset is the most common and typical one of various single event effects caused by space radiation, and soft errors are generated after the logic circuit generates single event upset. It is necessary to address this by fault tolerant design.
Rapid development of artificial life sciences has been in progress in the 20 th century, and various bionic structures and systems have been continuously developed. The electronic system designed based on the bionic hardware technology can autonomously and dynamically change the structure and the behavior of the electronic system to adapt to the living environment of the electronic system according to the change of the external environment in a reconstruction mode, the embryo type bionic hardware (often called embryo hardware Embryonics Hardware and EH for short) is a novel reconfigurable electronic hardware type inspired by the growth and development mechanism of multicellular organism tissues, has the self-repairing capability of distributed autonomous control, has the characteristics of self-organization, self-adaption, self-repairing and the like, and has higher reliability and stronger fault tolerance capability based on the electronic system designed by the embryo hardware technology. At present, embryo hardware research focus is hardware architecture and self-test and self-repair strategies. The method is characterized in that a multicellular organism structure of a natural organism is simulated, embryo hardware adopts a two-dimensional array structure and a distributed control mode, cells with the same structure are used for constructing a two-dimensional electronic array, circuit functions are mapped into each electronic cell, and due to the fact that the cell structures are the same, when working cells fail, the functions of the working cells can be replaced by redundant cells, and a repairing process is carried out by detecting the failure of the cells by the cells themselves and replacing the cells by the distributed autonomous control cells.
The improvement of the reliability of the electronic equipment is an important research direction of the design of the hardware circuit system in the aerospace environment, and along with the wider application of the reconfigurable electronic hardware in the aerospace field, the research significance of the high-reliability design method of the aerospace electronic system based on the embryo hardware technology is more and more remarkable. Electronic system faults in an aerospace environment have the characteristics of randomness and diversity, and can be mostly divided into transient faults and permanent faults, wherein the proportion of the transient faults is higher.
At present, embryo hardware design is mainly aimed at permanent faults, only a small number of researchers conduct instantaneous fault self-repairing method research, and only the validity of a specially designed circuit can be evaluated, the reliability of the proposed self-repairing method is evaluated, a reliability model capable of reflecting the influence of the instantaneous faults is lacking, namely, the reliability model only considers the permanent faults, and the influence on an electronic system for evaluating the space environment is particularly prominent because of larger inconformity with the actual situation.
Reliability is an important indicator for evaluating whether a system is valid and reliable. The reliability analysis of embryo hardware is carried out by taking a single cell removal mechanism as an object to carry out reliability modeling mainly, wherein the reliability analysis of the single cell removal strategy can be effectively carried out by the modeling method, but only permanent faults are considered for cell faults in the model. For a system with permanent faults and transient faults, even with a higher proportion of transient faults, the model has a larger error, and transient fault cells have the characteristic of being reusable after self-repairing, and the reuse capability cannot be represented in the model modeled by the permanent faults. Therefore, a new reliability modeling method capable of embodying transient fault characteristics is required.
Disclosure of Invention
The invention aims to: in order to make up for the defects in the prior art, the invention provides a reliability modeling and design guidance method for embryo hardware cell reuse strategies, which is based on cell fault characteristics in an aerospace environment, introduces fault types and transient fault proportion parameters by combining the cell reuse strategies on the basis of the traditional reliability modeling method, establishes a new reliability model for the embryo hardware cell reuse strategies, and summarizes indexes capable of evaluating the system design quality under different cell array scales and transient fault proportion conditions by analyzing the reliability model, and provides a guidance method for realizing design optimization according to the indexes.
In order to solve the technical problems, the invention provides a reliability modeling method for an embryo hardware cell reuse strategy, which is based on the cell reuse strategy and adds transient fault proportion parameters, establishes a reliability model for evaluating the cell utilization rate at different transient fault ratios by taking reliability as an optimization target, and can guide the optimal design of a cell array according to the reliability model.
The reliability model comprises a row reliability model, an array reliability model and an array reliability;
the row reliability R h The (t) model is:
wherein M 'represents the total number of cells in the equivalent row, M' =m+a, a is the number of repairable times in the cell row, and a=f 0 (M, k, δ), k being the in-line working cell duty ratio, k=m/M, M being the initial in-line total number of cells, δ being the instantaneous failure proportion parameter, indicating the probability that the failure type of any cell in the embryo array is an instantaneous failure when it fails, the probability that the failure type is a permanent failure is 1- δ, the failure that cannot be repaired by reconfiguration is all regarded as a permanent failure, M being the initial in-line working cell number, and the failure rate of cells being λ;
the array reliability R (t) model is as follows:
wherein N is the number of working cell lines, and the total number of cell lines in the array is N;
the reliability MTTF model is as follows:
the method for calculating the repairable times a in the cell line in the line comprises the following steps: and randomly determining the position of the faulty cells in the cell row, and simultaneously determining the cell type according to the instantaneous fault proportion parameter, and if the instantaneous faulty cells or idle cells are arranged at the right end of the cells, adding 1 to the repairable times a in the row until the randomly generated faulty cells are all working cells or permanent faulty cells to the right.
The invention also discloses a design guidance method for reliability modeling, which comprises the following steps:
s1: cell array in-line repair times a, array reliability MTTF model and average reconstruction under cell reuse strategyComparing the time with the value under the single cell removal strategy to obtain the corresponding cell array repair frequency promotion degree P a Degree of reliability improvement P M Average reconstruction time degree of rise P T The average reconstruction time represents the reconstruction time required by the primary fault repair in the row;
s2: mapping of cell array P a ，P M ，P T Curve, defining repair times lifting degree P a Higher than average reconstruction time degree of rise P T Is defined as a repair capability improving region, and a reliability improving degree P is defined M Higher than average reconstruction time degree of rise P T The interval of (2) is a reliability improving area;
s3: taking the repair capability lifting area and the reliability lifting area as quantization indexes of the reliability model optimization design;
s4: the influence of different array scales and different working cell proportions on a repair lifting area and a reliability lifting area of a cell array is analyzed, and the influence is used as a basis to provide an optimal design guidance for the embryo electronic cell layout under different space environments;
s5: the actual design can determine instantaneous fault proportion parameters according to the actual working conditions, and then can determine the optimal cell array scale and working cell proportion by taking the maximum repair times and the minimum time consumption as the standard according to the analysis result of the reliability model, thereby realizing the optimization design guidance.
The said
The said
The saidWherein t is clk The clock is reconstructed for the system.
The beneficial effects are that: compared with the prior art, the reliability model provided by the invention solves the problem of instantaneous fault modeling and can conveniently analyze the variation trend of system reliability under different instantaneous fault proportions. The three parameters of the instantaneous fault proportion of cells, the cell array scale and the working cell proportion are abstracted, reliability analysis is carried out, and the optimization design guidance can be realized according to the maximum repair times and the minimum time consumption standard. The invention integrates three important indexes of the system repair capability lifting degree, the system reliability lifting degree and the average reconstruction time lifting degree, and provides two optimization standards from the aspects of improving the system repair performance, the system reliability and reducing the array repair time consumption: and the repair capability improving region and the reliability improving region realize layout optimization of the embryo electronic cells in different space environments according to the standard.
Drawings
Fig. 1 is a- δ relationship diagram when m=200, k=0.8;
fig. 2 is a graph of a-delta relationship when k=0.8, where M is different;
fig. 3 is a graph of a-delta relationship when m=200 and k is different;
fig. 4 is a MTTF- δ relationship graph when m=200, k=0.8;
fig. 5 is a MTTF- δ relationship graph when k=0.8, where M is different;
fig. 6 is a MTTF-delta relationship graph when m=200, k is different;
fig. 7 is T when m=200, k=0.8 W -delta relationship diagram;
fig. 8 is T when k=0.8, where M is different W -delta relationship diagram;
fig. 9 is T when m=200, k is different W -delta relationship diagram;
fig. 10 is P when m=200 and k=0.8 a ，P M ，P T A curve as a function of delta;
fig. 11 shows P when k=0.8, unlike M a ，P T A curve as a function of delta;
fig. 12 shows that P is different from M when k=0.8 M ，P T A curve as a function of delta;
FIG. 13 is a schematic illustration of a deviceWhen m=200 and k is different, P a ，P T A curve as a function of delta;
fig. 14 is P when m=200 and k is different M ，P T A curve as a function of delta;
Detailed Description
The inventive process is further illustrated below in conjunction with examples.
A reliability modeling and design guiding method for embryo hardware cell reuse strategy aims at the problems that embryo hardware transient faults are high in occurrence and the existing reliability model only reflects permanent faults in different space environments, a new reliability model is built for the cell reuse strategy design method, an optimal design standard is given through reliability analysis, and then a guiding method for determining optimal design according to the standard is obtained.
The faults of embryo electronic hardware can be divided into transient faults and permanent faults, while the current reliability model only reflects the permanent faults, so that the actual conditions of the aerospace environment are accurately reflected, and transient fault parameters are required to be added in the reliability modeling process. The cells with transient faults in embryo hardware can be repaired by reconfiguration, namely the cells with transient faults can be removed firstly, then the cells are subjected to transparentization treatment (when the cells are added into a cell array as redundant idle cells and are reconfigured when the cells are used for repairing and replacing faulty cells next time), compared with a direct permanent removal strategy of permanent faulty cells, the number of cells which can be used for repairing and replacing is increased by adopting a cell reuse strategy, the number of repairable times is correspondingly increased, which is equivalent to the number of redundant cells increased, and the reliability of the array can be greatly improved. The invention aims to reflect the characteristics of the cell reuse strategy on the basis of the traditional reliability model and establish a new reliability model.
Inspired by the configurable repair characteristic of the transient fault cells, the invention abstracts the transient fault cells after the transparentization treatment into redundant idle cells, so that the equivalent total idle cell number in the cell line is increased, and the value of the equivalent total idle cell number is equal to the sum of the redundant idle cell number in the array line and the transient fault cell number after the transparentization treatment, thereby reflecting the reuse effect of the transient fault cells. A method for abstracting and modeling reusability of transient fault cells by using equivalent idle cell numbers is provided.
1. Reliability modeling
The reliability model comprises a row reliability model, an array reliability model and an array reliability model; the establishment of the reliability model comprises the following steps:
s1-1: according to a cell reuse strategy, calculating repairable times a in a cell line in combination with the probability of transient faults (transient fault proportion delta) of cells in the array;
s1-2: taking a as the new idle cell number in the row, and the equivalent cell scale in the row is M' =m+a; m and M respectively represent the total number of cells and the number of working cells in the initial row; m' represents the total number of cells in the equivalent row;
s1-3: establishing row reliability R according to the total number M' of cells in the equivalent post row, the working cell number M in the initial row and the cell failure rate lambda h (t) model:
s1-4: according to the working cell line number N, the array total cell line number N, the line reliability R h (t) establishing a reliability R (t) and reliability MTTF model of the array:
2. reliability analysis and parameter calculation
In order to obtain the guiding method of the optimal design, various parameters and indexes are required to be determined according to the working environment of the system and the design reality. The analysis of parameters and indexes is performed on the reliability model, and the cell array with the number of embryo cells being NxM and the number of working cells being NxM is taken as an example, and the analysis method is used for the explanation.
(1) Calculating in-line reliability
Setting transient fault proportion parameters delta according to different probabilities of transient faults of cells in different space environments: indicating the probability that any cell in the embryo array will fail, the failure type will be transient. Analyzing the relation between repairable times and transient fault proportion in a row: a-delta change curve.
Let k be the intra-row working cell duty cycle, k=m/M. Available a=f 0 The (M, k, δ) function represents the influencing factor of a. The calculation method of a is as follows: and randomly determining the position of the faulty cells in the cell row, simultaneously determining the cell type according to the transient fault proportion delta, and if the transient faulty cells or idle cells exist at the right end of the cells, adding 1 to the number of in-row repair times a until the randomly generated faulty cells are all working cells or permanent faulty cells to the right. Since the location of faulty cells was random, the statistical a value was averaged over 100 experiments.
Parameter analysis is carried out on repairable times and instantaneous fault proportion parameters (delta value is from 0 to 1) in the row, and the three conditions are as follows:
(1) fixed array scale, fixed working cell ratio: taking m=200, k=0.8, then a=f 0 (200,0.8,δ)；
Fig. 1 shows a=f 0 (200,0.8, δ). Under the cell reuse strategy, the number of repairable times a in a row increases with the increase of the transient fault proportion delta. For traditional single cell removal strategies, the number of repairable times in a row is the same as the number of redundant free cells, independent of δ.
(2) Different array scales, fixed working cell ratios: taking k=0.8, m 20,50,100,150,200, then a=f 0 (M,0.8,δ)；
Fig. 2 shows a=f 0 Analysis results of (M, 0.8, delta). With the expansion of the array scale, the number of idle cells in a row is increased, and the repairable times a in the row is increased; at each cell array scale, the number of repairable times a in the row increases with delta; and as delta increases, the difference of repair times in the cell array row under adjacent scale gradually increases, namely, the larger the array scale is, the repair is performedThe more the number of times increases.
(3) Fixed array scale, different working cell ratios: taking m=200, k is 0.5,0.6,0.7,0.8,0.9, respectively, the repair capability function is a=f 0 (200,k,δ)；
Fig. 3 shows a=f 0 Analysis results of (200, k, δ). The number of idle cells changes with the proportion of working cells, and the number of idle cells is smaller when k is larger, so that the number of repairable times in the line is reduced, but as delta is increased, the difference of the number of repairable times in the line under the proportion of adjacent working cells is also increased.
Reliability of calculation of
The reliability is analyzed as a function of the transient fault proportion. Performing repair frequency function a=f on cell array 0 (M, k, delta) into a new reliability model, an array reliability model for the transient fault proportion parameter delta can be obtained:
row reliability:
array reliability:
array reliability:
available mttf=g 0 The (M, k, N, N, delta, lambda, t) function represents the influencing factor of the MTTF. Considering that the cell reuse strategy is mainly used for reconstructing transient faults in a row, for the convenience of calculation, the column parameters are selected to be N=M, n=m, and the cell failure rate lambda=1×10 -6 The example analysis is performed/h, and thus the reliability function of the cell array can be expressed as: mttf=g 0 (M,k,N,n,δ,λ,t)＝g 1 (M,k,δ,t)。
Based on the cell reuse strategy and the traditional single cell removal strategy, the array reliability at different delta is calculated respectively.
(1) Fixed array scale, fixed working cell ratio: fig. 4 shows mttf=g 1 (200,0.8,δAnalysis results of t). The reliability MTTF of the cell array is consistent with the change rule of the cell repair times a. The reliability of the array under the cell reuse strategy is higher.
(2) Different array scales, fixed working cell ratios: fig. 5 shows an array reliability function mttf=g 1 Analysis results of (M, 0.8, delta, t). When delta is smaller, the cell array scale has little effect on the array reliability; as δ increases, the larger the array size, the higher the reliability.
(3) Fixed array scale, different working cell ratios: fig. 6 shows mttf=g 1 Analysis results of (200, k, δ, t). Since the size of k directly affects the number of free cells, the array reliability for each working cell ratio increases as δ increases, and when δ is the same, the larger k the lower the cell array reliability.
Calculating average reconstruction time
The average reconstruction time is the time penalty required for array reconstruction under a cell reuse strategy. Let the array average reconstruction time be T W The method represents the reconstruction time required by repairing one fault in a row, and the reconstruction time of each repair is different due to different types and positions of faults, and the calculation process takes an average value of 100 statistics.
Usable T W ＝h 0 (X,δ)·t clk The function represents T W Wherein X is the number of permanent faulty cells between each newly generated fault in the row and its spare cells (or transient faulty cells) that can be repaired (each permanent fault resulting in an unsuccessful self-repair operation), t clk The clock is reconstructed for the system, always taking 1. The size of X is related to the array scale, the proportion of working cells. Thus, T can be used W ＝h 0 (X,δ)·t clk ＝h 1 (M,k,δ)·t clk And (3) representing. Average reconstruction time T W The specific calculation and analysis process of (a) is as follows:
(1) fixed array scale, fixed working cell ratio: FIG. 7 shows T W ＝h 1 (200,0.8,δ)·t clk Is a result of the analysis of (a). The average reconstruction time under the condition of realizing the cell reuse is greatly higher than that of the traditional single cellCell removal strategy (1 failure requires only 1 reconstruction clock cycle, independent of delta). For cell reconstruction strategy, at delta. Epsilon.0, 0.8]Interval, average reconstruction time T W After the maximum value is reached, as delta is increased, the number of times of repairing processes is reduced as permanent faults are less, and the tie reconstruction time is rapidly reduced. When δ=1, the average reconstruction time T W ＝1。
(2) Different array scales, fixed working cell ratios: FIG. 8 shows T W ＝h 1 (M,0.8,δ)·t clk Is a result of the analysis of (a). As the array scale increases, the average reconstruction time of the array increases due to the same delta as the total number of cells increases.
(3) Fixed array scale, different working cell ratios: FIG. 9 shows T W ＝h 1 (200,k,δ)·t clk Is a result of the analysis of (a). The trend of the average reconstruction time curve of the cell array with different working cell ratios is unchanged, but the larger the working cell ratio is, the shorter the average reconstruction time is.
3. Optimization design guiding method
The number of cell array in-row repair a, array reliability MTTF and average reconstruction time T are analyzed W Aiming at a cell reuse strategy, an optimization standard is set from the angles of enhancing the repair capability of the system, improving the reliability of the array and reducing the average reconstruction time, and guidance is provided for the optimal layout mode of cells in different space environments.
(1) Determining design optimization criteria
Comparing the intra-row repair times, array reliability and average reconstruction time of the cell array under the cell reuse strategy with the values under the traditional single-cell removal strategy to obtain the corresponding degree P of improvement of the repair times of the cell array a Degree of reliability improvement P M Average reconstruction time degree of rise P T And (5) an index.
P a ，P M ，P T Is defined as follows:
equal to twoUnder seed strategy, the ratio of the number of times of self-repair in the rows in the array; under the traditional single cell removal strategy, the repairable times in the row are always equal to the number of idle cells in the row.
Equal to the ratio of array reliability under two strategies; the state of the conventional single cell removal strategy is the same as the state of δ=0 under the cell reconstitution strategy.
The ratio of the reconstruction time required for repairing 1 fault under two strategies is equal to that of the two strategies; under the traditional single cell removal strategy, the reconstruction time for 1 self-repair is 1 clock cycle.
(2) Establishing P a ，P M ，P T The model is analyzed and design optimization criteria are given.
FIG. 10 shows the P of the cell array at a fixed working cell fraction for a fixed array scale M=200, m=160 a ，P M ，P T Curve delta epsilon [0.96,1 ]]At the time P a Above P T Defined as repair capability promotion area Q a The repair time increase rate is higher than the average reconstruction time consumption increase rate; delta epsilon [0.97,1 ]]At the time P M Above P T Defined as a reliability enhancement region Q M The reliability improvement rate is higher than the average reconstruction time consumption increase rate. Therefore, the design guidance can be optimized according to the two areas, the design falling in the two areas can obtain better performance by adopting a cell reuse strategy.
(3) The design optimization guiding method comprises the following steps: the designer calculates and obtains two lifting area indexes according to the actual design environment, the array scale and the duty ratio of the working cells, and then determines which strategy is adopted for design, or optimizes specific indexes according to the actual situation.
(4) Design instruction example
Will P a ，P M ，P T Cross analysis model graph of (c) is divided into P a ，P T Graph and P M ，P T And (5) respectively carrying out comparison analysis.
(1) FIGS. 11 and 12 show P at the same working cell ratio for different array sizes a ，P T Graph and P M ，P T A figure; taking k as 0.8 for analysis, and along with the expansion of array scale, obtaining interval Q a And Q M Are gradually decreasing. It is shown that as array scale increases, the proportion of transient faults in the spatial environment embodying the superiority of cell reuse strategies needs to be higher. Of course, whether to adopt the ratio comparison of the repair times improvement rate and the reliability improvement rate to the average reconstruction time improvement rate as the judgment basis or not needs to consider the practical application demand factors, so the practical application optimization interval is larger.
(2) FIGS. 13 and 14 show P at different ratios of working cells for the same array scale a ，P T Graph and P M ，P T A figure; from the graph, the interval Q increases with the proportion of working cells while the array scale m=200 remains unchanged a Sum interval Q M Gradually increasing, the advantages of the cell reuse strategy are more obvious.
Based on the above description, the reliability modeling and design guidance method of the present invention provides a system repair capability improvement degree P based on transient fault proportion a Degree of improvement P of system reliability M Average reconstruction time degree of rise P T Based on cell reuse strategy, to obtain repair capacity improving region Q of cell array a And a reliability improving region Q M Two design optimization indexes are used for analyzing the optimal repair capacity interval Q of the cell array by different array scales and different working cell proportions a And an optimal reliability interval Q M And the influence of the electron cell is used as a basis to provide an optimal design guidance for the embryo electron cell layout under different space environments.
The above examples are not intended to limit the present invention in any way, and all reliability models obtained by the above reliability modeling method and embryo systems obtained under other embryo cell scale arrays using the above design guidance scheme fall within the scope of the present invention.
The invention protects an analysis method and a design guidance method for reliability modeling, and a corresponding reliability model can be established through analysis of the method, so that environmental adaptability of different cell removal strategies under different space environments can be analyzed, and influence of the space environments on the repair performance of an embryo electronic cell array and the reliability of a system can be analyzed. The design guidance method provided by the invention can provide design guidance for the embryo electronic cell layout in different space environments, can be used as an evaluation basis for whether the design is feasible in a specific environment, and has important significance for application and practical design verification. The reliability modeling method and the reliability model provided by the invention are not only the establishment of the reliability model, but also the key is to grasp the key numerical index of the reliability according to the characteristics of the fault tolerance strategy of the current embryo self-repairing hardware and analyze the influence of the space environment on the embryo array structure and the fault tolerance capacity, so as to further provide a design guidance method. Although the embryo electronic array under different space environments gives the optimal repair capability interval Q a And an optimal reliability interval Q M The design guidance method provided by the invention has general applicability, and examples of the invention show a selection method of a cell array working environment and a layout optimization method of a cell array when the array system is kept in a high repair state and an optimal repair state under the guidance of the design method.
The above is merely a reliability modeling and design guidance method of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the present invention, and such modifications and adaptations are intended to be comprehended within the scope of the invention. The specific embryo circuit array selection and use of cell removal strategies are not within the scope of the present invention, but different forms of cell circuit arrays and cell removal strategies, the reliability models and optimization metrics obtained by the reliability modeling analysis methods and design guidance methods herein are limited by the rights of the present invention.
Claims (3)
1. A cell array optimization design guidance method is characterized in that: the method comprises the following steps:
s1: comparing the cell array intra-row repair times a, the array reliability model MTTF and the average reconstruction time under the cell reuse strategy with the value under the single cell removal strategy to obtain the corresponding cell array repair times lifting degree P a Degree of reliability improvement P M Average reconstruction time degree of rise P T The average reconstruction time represents the reconstruction time required by the primary fault repair in the row;
s2: mapping of cell array P a ，P M ，P T Curve, defining repair times lifting degree P a Higher than average reconstruction time degree of rise P T Is defined as a repair capability improving region, and a reliability improving degree P is defined M Higher than average reconstruction time degree of rise P T The interval of (2) is a reliability improving area;
s3: taking the repair capability lifting area and the reliability lifting area as quantization indexes of the reliability model optimization design;
s4: the influence of different array scales and different working cell proportions on a repair capacity improving region and a reliability improving region of the cell array is analyzed, and an optimal design guidance is provided for embryo electronic cell layout under different space environments according to the influence;
s5: determining instantaneous fault proportion parameters according to actual working conditions, and determining optimal cell array scale and working cell proportion by taking maximum repair times and minimum time consumption as standards according to analysis results of a reliability model so as to realize optimization design guidance;
the reliability model comprises a row reliability model, an array reliability model and an array reliability model;
the row reliability model is expressed as:
wherein f 0 (M, k, δ) represents the number of cell array in-line repairs, a=f 0 (M, k, δ), k being the in-line working cell fraction, k=m/M, M being the total number of cells in the initial line, δ being the instantaneous failure proportion parameter, representing the probability that any cell in the embryo array fails, that its failure type is instantaneous failure, M being the number of working cells in the initial line, λ being the failure rate of the cells;
the array reliability model is expressed as:
wherein N is the number of working cell lines, and the total number of cell lines in the array is N;
the array reliability model is expressed as:
2. the cell array optimization design guidance method according to claim 1, wherein:
the said
The said
The saidWherein t is clk Reconstructing a clock for the system; t (T) W Represents the average reconstruction time, T W ＝h 1 (M,k,δ)·t clk Representing T W Influence factors and calculation methods of the system.
3. The cell array optimization design guidance method according to claim 1, wherein: the method for calculating repairable times a in the cell array row comprises the following steps: and randomly determining the position of the faulty cells in the cell row, and simultaneously determining the cell type according to the instantaneous fault proportion parameter, and if the instantaneous faulty cells or idle cells are arranged at the right end of the cells, adding 1 to the repairable times a in the row until the randomly generated faulty cells are all working cells or permanent faulty cells to the right.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811269642.6A CN109635315B (en) | 2018-10-29 | 2018-10-29 | Reliability modeling and design guiding method oriented to embryo hardware cell reuse strategy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811269642.6A CN109635315B (en) | 2018-10-29 | 2018-10-29 | Reliability modeling and design guiding method oriented to embryo hardware cell reuse strategy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109635315A CN109635315A (en) | 2019-04-16 |
| CN109635315B true CN109635315B (en) | 2023-12-08 |
Family
ID=66066765
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811269642.6A Active CN109635315B (en) | 2018-10-29 | 2018-10-29 | Reliability modeling and design guiding method oriented to embryo hardware cell reuse strategy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109635315B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110188453B (en) * | 2019-05-27 | 2021-01-15 | 中国人民解放军国防科技大学 | Bionic self-repairing hardware dynamic layout optimization method and system |
| CN110991128A (en) * | 2019-12-03 | 2020-04-10 | 南京航空航天大学 | Cell array circuit-based FPGA self-repairing structure and fault-tolerant method thereof |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2837984A2 (en) * | 2013-08-05 | 2015-02-18 | Uptime Engineering GmbH | Process to optimize the maintenance of technical systems |
| CN104392149A (en) * | 2014-12-11 | 2015-03-04 | 南京航空航天大学 | Cellular array structure capable of improving reliability of embryonal bionic self-repairing reconfigurable hardware and layout method for cellular array structure |
-
2018
- 2018-10-29 CN CN201811269642.6A patent/CN109635315B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2837984A2 (en) * | 2013-08-05 | 2015-02-18 | Uptime Engineering GmbH | Process to optimize the maintenance of technical systems |
| CN104392149A (en) * | 2014-12-11 | 2015-03-04 | 南京航空航天大学 | Cellular array structure capable of improving reliability of embryonal bionic self-repairing reconfigurable hardware and layout method for cellular array structure |
Non-Patent Citations (3)
| Title |
|---|
| Self-healing strategy for transient fault cell reutilization of embryonic array circuit;张砦 等;《2018 NASA/ESA Conference on Adaptive Hardware and Systems (AHS)》;20180809;第225-232页 * |
| 具有故障细胞的胚胎电子阵列上目标电路评估;朱赛等;《兵工学报》;20161115(第11期);全文 * |
| 基于可靠性分析的胚胎硬件容错策略选择方法;张砦等;《系统工程理论与实践》;20130115(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109635315A (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220164285A1 (en) | Compensating for dram activation penalties | |
| CN109635315B (en) | Reliability modeling and design guiding method oriented to embryo hardware cell reuse strategy | |
| EP0912979A1 (en) | Semiconductor memory tester with redundancy analysis | |
| CN111143764B (en) | Distribution network reliability assessment method with complex structure based on diffusion theory | |
| CN111784116B (en) | Method and system for evaluating elastic capability of power distribution network under multi-energy coupling | |
| CN101630534B (en) | Method and device for testing reliability of nonvolatile memories | |
| CN115310048B (en) | Method and system for calculating repair probability of equipment in expected time | |
| Kline et al. | Sustainable fault management and error correction for next-generation main memories | |
| Ma et al. | BlockHammer: Improving flash reliability by exploiting process variation aware proactive failure prediction | |
| CN106777719B (en) | The fault detection method of the failure detector of duplication redundancy | |
| Wang et al. | Real-time fault repair scheme based on improved genetic algorithm | |
| CN117148117B (en) | Chiplet fault automatic detection and repair method and system | |
| Lu et al. | Synergistic reliability and yield enhancement techniques for embedded SRAMs | |
| CN111241685B (en) | Reliability optimization design method for lithium battery pack system | |
| CN115270078B (en) | Method and system for calculating average repair time of electromechanical equipment | |
| Condia et al. | Protecting GPU's microarchitectural vulnerabilities via effective selective hardening | |
| CN115437893A (en) | Cluster analysis-based integrated circuit single event effect soft error simulation method | |
| CN107895090B (en) | Mathematical description method of embryo electronic system | |
| Cha et al. | Characterization of granularity and redundancy for SRAMs for optimal yield-per-area | |
| Lin et al. | Highly reliable and low-cost symbiotic IOT devices and systems | |
| Kumar et al. | Supervised machine leaning based anomaly detection and diagnosis in grid connected photovoltaic systems | |
| JP2010113751A (en) | Method and device of redundancy arithmetic operation, and memory test device | |
| CN109669803A (en) | For reducing the method and apparatus of the soft error rate of ECC memory | |
| CN115421967B (en) | Method and system for evaluating storage abnormal risk point of secondary equipment | |
| KR102583916B1 (en) | Scan correlation-aware scan cluster reordering method and apparatus for low-power testing |
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 |