A data storage method and data storage system for a memory chip

Abstract

The present invention discloses a data storage method and a data storage system of a storage chip, comprising: selecting reference data and sample data in target storage data and analyzing them to obtain a data similarity value; comparing the data similarity value with a data similarity threshold, and if the data similarity value is greater than or equal to the data similarity threshold, classifying the reference data and the sample data into the same classified data group; obtaining remaining storage space, judging whether the remaining storage space is less than the remaining storage space threshold, and if so, analyzing the data in the classified data group to obtain a data group key value; identifying a non-key data group and generating a compression optimization signal; judging based on the compression optimization signal whether the remaining storage space of the system after compression processing is less than the remaining storage space threshold, and if so, analyzing the non-high-access-frequency data to obtain a compression priority value; and optimizing and compressing the non-high-access-frequency data according to the compression priority value.

Description

Technical Field

[0001] The present invention belongs to the technical field of data storage, and in particular to a data storage method and a data storage system for a storage chip. Background Art

[0002] As electronic products become more popular, people have more and more demands for the intelligence of electronic products. Accordingly, the requirements for chip performance and chip storage data stability in electronic products are also getting higher and higher. Storage chips in electronic products need to efficiently manage data and ensure performance and stability to meet users' needs for intelligent experience. In particular, in terms of data storage and processing, storage chips need to have intelligent optimization capabilities, adjust storage strategies, improve data access efficiency, ensure stable system operation, and thus improve user experience. Current storage technology is difficult to meet the growing data management needs.

[0003] To solve this problem, the present invention proposes a data storage method and a data storage system for a storage chip. The data is stored in a classified manner. When the remaining storage space is insufficient, each type of data group is analyzed. Based on the analysis results, non-critical data groups are compressed and optimized to ensure that critical data groups are not affected. At the same time, the performance and stability of the storage chip are effectively improved, ensuring that the remaining storage space in the system meets the daily needs of users. Summary of the invention

[0004] An object of the present invention is to provide a data storage method and a data storage system for a memory chip to solve at least one of the above-mentioned problems in the prior art.

[0005] In a first aspect, the present invention provides a data storage method of a memory chip, comprising:

[0006] Step 1: Select reference data and sample data from the target storage data, analyze the data labels and data volumes of the reference data and sample data, and obtain data similarity values;

[0007] Step 2: Compare the data similarity value with the data similarity threshold. If the data similarity value is greater than or equal to the data similarity threshold, store the reference data and the sample data in the same classification data group;

[0008] Step 3: After the data storage is completed, the remaining storage space of the system is obtained; the remaining storage space is compared with the remaining storage space threshold, and if the remaining storage space is less than the remaining storage space threshold, the historical access frequency of the data in the classified data group is analyzed to obtain the key value of the data group;

[0009] Step 4: Compare the data group key value with the data group key threshold, and determine whether the classified data group is a non-key data group based on the comparison result. If so, generate a compression optimization signal;

[0010] Step 5: Based on the compression optimization signal, perform data compression processing on the non-critical data group; determine whether the remaining storage space after the compression processing is less than the remaining storage space threshold; if so, analyze the storage duration and access times of the non-high-access-frequency data in the critical data group to obtain a compression priority value;

[0011] Step 6: Sort the non-high access frequency data from high to low according to the compression priority value to obtain a compression priority sequence; based on the compression priority sequence, compress the non-high access frequency data in sequence from small to large according to the serial number; when the remaining storage space reaches the remaining storage space threshold, stop compression.

[0012] In a second aspect, the present invention provides a data storage method of a memory chip, comprising:

[0013] Similarity parameter acquisition module: selects reference data and sample data from the target storage data, analyzes the data labels and data volumes of the reference data and sample data, and obtains data similarity values;

[0014] Similarity parameter analysis module: compares the data similarity value with the data similarity threshold. If the data similarity value is greater than or equal to the data similarity threshold, the reference data and the sample data are stored in the same classification data group.

[0015] Key data parameter acquisition module: after completing data storage, obtain the remaining storage space of the system; compare the remaining storage space with the remaining storage space threshold; if the remaining storage space is less than the remaining storage space threshold, analyze the historical access frequency of the data in the classified data group to obtain the key value of the data group;

[0016] Key data parameter analysis module: compares the key value of the data group with the key threshold of the data group, and determines whether the classified data group is a non-key data group based on the comparison result. If so, a compression optimization signal is generated;

[0017] Compression parameter acquisition module: based on the compression optimization signal, the non-critical data group is compressed; after the compression process, the remaining storage space is determined to be less than the remaining storage space threshold; if so, the storage duration and access times of the non-high-frequency access data in the critical data group are analyzed to obtain the compression priority value;

[0018] Compression parameter analysis module: Sort the non-high access frequency data from high to low according to the compression priority value to obtain the compression priority sequence; based on the compression priority sequence, compress the non-high access frequency data in sequence from small to large according to the serial number; when the remaining storage space reaches the remaining storage space threshold, stop compression.

[0019] Beneficial effects of the present invention:

[0020] 1. The technical solution of the embodiment of the present invention is: classify and store the data to be stored and ensure efficient classification of the data through a similar parameter analysis module, thereby effectively improving storage efficiency; by classifying and storing different types of data in different storage areas or media, the risk of data loss due to the failure of a single storage area or medium can be avoided; by classifying the data, it is more convenient to retrieve, query, back up and restore the data; both the availability and integrity of the stored data are guaranteed.

[0021] 2. The technical solution of the embodiment of the present invention is: by real-time monitoring of the remaining storage space of the system, the problem of insufficient remaining storage space of the system is discovered in time; to avoid data loss due to insufficient remaining storage space of the system, the classified data group is analyzed to determine whether the classified data group is a non-critical data group, and if so, the non-critical data group is compressed and optimized; through compression optimization, storage space is released to ensure storage security and access efficiency of critical data groups.

[0022] 3. The technical solution of the embodiment of the present invention is: by real-time monitoring of the remaining storage space of the system, after the system completes the identification and compression optimization of the non-critical data group, it is determined whether the remaining storage space of the system meets the system operation requirements. If not, the critical data group is analyzed to identify the non-high-access frequency data in the critical data group, and the compression priority value is determined according to the storage length and the historical access times of the data, and the data is arranged according to the compression priority value of the non-high-access frequency data to obtain a compression priority sequence; based on the compression priority sequence, the non-high-access frequency data is orderly compressed; by effectively identifying the non-critical data in the system storage data, secondary compression is performed when the remaining storage space is insufficient, so as to further release the storage space of the system, and ensure that the critical data can be read normally and the system can run stably. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 This is a flowchart of the steps of a data storage method of a memory chip provided in the first embodiment of the present invention;

[0025] Figure 2 It is a flow chart of obtaining data similarity values ​​in a data storage method of a memory chip provided in the first embodiment of the present invention;

[0026] Figure 3This is a flow chart of a data storage method of a memory chip for obtaining a key value of a data group provided by the second embodiment of the present invention;

[0027] Figure 4 It is a flow chart of a data storage method for a memory chip for obtaining a compression priority value provided by a third embodiment of the present invention;

[0028] Figure 5 It is a module schematic diagram of a data storage system of a memory chip provided in a fourth embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention. Example

[0030] like Figure 1 and Figure 2 As shown, a data storage method of a memory chip provided by an embodiment of the present invention specifically includes the following steps:

[0031] Step 1: Select reference data and sample data from the target storage data, analyze the data labels and data volumes of the reference data and sample data, and obtain data similarity values;

[0032] It should be noted that the data volume of the target storage data indicates the data size of the data to be stored;

[0033] S11: Sort the target storage data according to the time sequence of the system reading the data to obtain a storage data sequence, marked as T, where ; Store the data in a sequence, data Marked as reference data, data Marked as sample data;

[0034] Obtain data labels of reference data, recorded as the number of reference data labels;

[0035] Based on any sample data, obtain the data label of the sample data, which is recorded as the number of sample data labels;

[0036] It should be noted that the number of reference data labels of the reference data and the number of sample data labels of the sample data are not unique;

[0037] S12: by traversing the number of sample data labels of each sample data, calculating the number of overlapping labels of each sample data and the reference data, the number of overlapping labels of each sample data and the reference data is obtained, which is recorded as the number of overlapping labels;

[0038] The number of overlapping labels is processed by ratio with the number of reference data labels to obtain a reference overlap ratio; the number of overlapping labels is processed by ratio with the number of sample data labels of the corresponding sample data to obtain a sample overlap ratio;

[0039] It should be noted that the purpose of obtaining the reference overlap ratio and the sample overlap ratio is to avoid a large difference between the number of reference data labels and the number of sample data labels, which may lead to deviations in the calculation results;

[0040] The reference coincidence ratio and the sample coincidence ratio are weighted summed to obtain the coincidence label ratio, which is marked as CH;

[0041] It should be noted that the purpose of the overlapped label ratio CH is to analyze the number of overlapped labels between the reference data labels and the sample data labels, and then evaluate the degree of the reference data type and the sample data type, so as to facilitate the subsequent classification and storage of the data;

[0042] S13: Obtaining the data volume of reference data and the data volume of sample data;

[0043] The reference data volume and the sample data volume are subjected to difference processing and the absolute value is taken to obtain the data volume difference; the ratio of the data volume difference to the reference data volume is recorded as the data volume difference ratio, marked as CB;

[0044] S14: By analyzing the overlapped label ratio CH and the data volume difference ratio CB, the formula , and obtain the data similarity value XS; where a1 and a2 are preset proportional coefficients;

[0045] It should be noted that the data similarity value XS is used to measure the similarity between the reference data and the sample data, which is convenient for the subsequent classification and storage of the data;

[0046] Step 2: Compare the data similarity value with the data similarity threshold. If the data similarity value is greater than or equal to the data similarity threshold, store the reference data and the sample data in the same classification data group;

[0047] S21: Compare the data similarity value with the data similarity threshold. The comparison process is as follows:

[0048] If the data similarity value is greater than or equal to the data similarity threshold, a similarity signal is generated, and the corresponding sample data is marked as similar sample data; the reference data and the similar sample data are classified into the same classification data group;

[0049] If the data similarity value is less than the data similarity threshold, a non-similar signal is generated; the corresponding sample data is marked as non-similar sample data; the reference data and the non-similar sample data are not classified into the same classification data group;

[0050] S22: After the reference data and similar sample data are classified into the same classification data group, the reference data is reselected from the remaining data, and similarity value analysis is continued until all sample data are classified; and the data is classified and stored according to the classification data group;

[0051] For example, in storing data sequence In, if For similar sample data, into the same classification data group; After being classified into the same classification data group, the remaining data are rearranged to obtain ,by For new reference data, For new sample data, similarity value analysis is continued until all sample data are classified;

[0052] The technical solution of the embodiment of the present invention is: by selecting reference data and sample data from the data to be stored, analyzing the number of overlapping labels and the data volume difference between the reference data and the sample data, and calculating the data similarity value; judging whether to classify the sample data and the reference data into the same data group based on the comparison of the data similarity value with a preset threshold, thereby realizing accurate classification and storage of data; classified storage makes the organization of data more orderly and facilitates the rapid location of required data; this helps to reduce data retrieval time and improve data access speed. Example

[0053] like Figure 1 and Figure 3 As shown, a data storage method of a memory chip provided by an embodiment of the present invention specifically includes the following steps:

[0054] Step 3: After the data storage is completed, the remaining storage space of the system is obtained; the remaining storage space is compared with the remaining storage space threshold, and if the remaining storage space is less than the remaining storage space threshold, the historical access frequency of the data in the classified data group is analyzed to obtain the key value of the data group;

[0055] S31: monitoring the remaining storage space of the system;

[0056] Obtaining the remaining storage space capacity, comparing the remaining storage space with a remaining storage space threshold, and if the remaining storage space is less than or equal to the remaining storage space threshold, generating an analysis signal, and compressing and optimizing the storage data, specifically;

[0057] S32: Based on any classification data set;

[0058] Obtain the historical access times and data storage duration of the data in the classified data group; perform ratio processing on the historical access times of the data and the data storage duration to obtain the access frequency value;

[0059] Compare the access frequency value with the preset access frequency threshold. If the access frequency value is higher than the threshold, mark the data as high access frequency data; otherwise, mark the corresponding data as non-high access frequency data.

[0060] S33: obtaining the number of high-access frequency data in the classified data group, performing ratio processing on the number of high-access frequency data and the total number of data in the classified data group to obtain a high-access data ratio, which is marked as SJ;

[0061] S34: obtaining access frequency values ​​of high access frequency data in the classified data group, summing and averaging the access frequency values ​​of all high access frequency data in the classified data group to obtain a high access frequency mean; performing ratio processing on the high access frequency mean and the standard high access frequency mean to obtain a high access frequency mean ratio, which is marked as FW;

[0062] It should be noted that if there is no high-access frequency data in the classification data group, the high-access frequency mean ratio FW takes a value of zero;

[0063] S35: Process the high access data ratio SJ and the high access frequency mean ratio FW, and use the formula , get the key value GJ of the data group; where b1 and b2 are preset coefficients;

[0064] It should be noted that the data group key value GJ reflects the importance and activity of the data group; analyzing the data group key value facilitates the evaluation of the importance of the data group and the formulation of subsequent compression optimization strategies for storage data to ensure that key data is stored first;

[0065] Step 4: Compare the data group key value with the data group key threshold, and determine whether the classified data group is a non-key data group based on the comparison result. If so, generate a compression optimization signal;

[0066] Compare the data group key value with the data group key threshold. The comparison process is as follows:

[0067] If the data group key value is greater than or equal to the data group key threshold, a retention signal is generated; and the corresponding data group is marked as a key data group;

[0068] If the data group key value is less than the data group key threshold, a compression optimization signal is generated; and the corresponding classified data group is marked as a non-key data group;

[0069] Based on the compression optimization signal, data compression processing is performed on the classified data group to release storage space;

[0070] The technical solution of the embodiment of the present invention is: analyzing the remaining storage space of the system, if the remaining storage space of the system is insufficient, analyzing the storage time and historical access times of the data in the classified data group, accurately identifying high-access frequency data, analyzing the importance and activity of the classified data group, obtaining the key value of the data group, formulating a compression optimization method according to the key value of the data group, giving priority to compressing the data group with a low key value of the data group, ensuring the reasonable allocation of the system storage space, avoiding insufficient remaining storage space in the system, resulting in unstable system operation; reasonably allocating the system storage space, effectively improving data storage efficiency, and ensuring rapid access to key data and stable operation of the system. Example

[0071] like Figure 1 and Figure 4 As shown, a data storage method of a memory chip provided by an embodiment of the present invention specifically includes the following steps:

[0072] Step 5: Based on the compression optimization signal, perform data compression processing on the non-critical data group; determine whether the remaining storage space after the compression processing is less than the remaining storage space threshold; if so, analyze the storage duration and access times of the non-high-access-frequency data in the critical data group to obtain a compression priority value;

[0073] S51: performing data compression processing on the non-critical data group based on the compression optimization signal;

[0074] It should be noted that in the data storage system of the memory chip, Huffman coding is used to compress data; Huffman coding is a variable-length coding method based on statistics, which allocates the code length according to the probability of the symbol appearing in the data; common symbols are assigned shorter codes, and uncommon symbols are assigned longer codes; Huffman coding can effectively reduce the overall size of the data while maintaining the integrity of the data;

[0075] S52: Compare the remaining storage space of the system after the data compression processing with the remaining storage space threshold. If the remaining storage space of the system after the data compression processing is less than the remaining storage space threshold, further analyze the non-high access frequency data in the key data group, specifically;

[0076] A1: Obtain the non-access duration of the non-high-access-frequency data in the key data group, perform a ratio process on the non-access duration and the data storage duration to obtain the non-access duration ratio, which is marked as WF;

[0077] A2: Obtain the access frequency value of the non-high access frequency data in the key data group, marked as AF;

[0078] A3: Process the non-visit duration ratio WF and the access frequency value AF, and use the formula , get the compression priority value YX; where c1 and c2 are preset proportional coefficients;

[0079] It should be noted that the compression priority value YX indicates the priority of data compression processing. The larger the compression priority value YX is, the higher the priority is. When the system has insufficient storage space, the data with the larger compression priority value YX is compressed first to maintain efficient and stable operation of the system.

[0080] Step 6: Sort the non-high-access-frequency data from high to low according to the compression priority value to obtain a compression priority sequence; based on the compression priority sequence, compress the non-high-access-frequency data in order from small to large according to the sequence number; when the remaining storage space reaches the remaining storage space threshold, stop compressing;

[0081] Obtain compression priority values ​​of all non-high-access-frequency data, sort the non-high-access-frequency data from high to low according to the compression priority values ​​YX of all non-high-access-frequency data, and obtain a compression priority sequence; obtain serial numbers of all non-high-access-frequency data, and compress the non-high-access-frequency data one by one according to the serial numbers from small to large;

[0082] Monitor the changes in the remaining storage space of the system in real time. If it is detected that the remaining storage space of the system has reached the remaining storage space threshold, stop compressing the remaining non-high-frequency access data. If it is detected that the remaining storage space of the system has not reached the threshold, continue to compress until the remaining storage space of the system reaches the remaining storage space threshold.

[0083] The technical solution of the embodiment of the present invention is: by judging whether the remaining storage space of the system after data compression processing is less than the remaining storage space threshold, if it is less, analyzing the non-high access frequency data in the key data group, analyzing the data storage duration and historical access times of the non-high access frequency data, and obtaining the compression priority value; sorting according to the compression priority value YX, generating a compression priority sequence, and compressing the non-high access frequency data in turn until the system storage space meets the standard; it is beneficial to identify and compress non-critical data when the remaining storage space of the system is insufficient, thereby increasing the remaining storage space of the system, improving the operating efficiency of the system, ensuring the reading and writing speed of key data, and leaving enough storage space for data updating. Example

[0084] like Figure 5 As shown, a data storage system of a memory chip provided by an embodiment of the present invention specifically includes the following modules:

[0085] Similarity parameter acquisition module: selects reference data and sample data from the target storage data, analyzes the data labels and data volumes of the reference data and sample data, and obtains data similarity values;

[0086] Similarity parameter analysis module: compares the data similarity value with the data similarity threshold. If the data similarity value is greater than or equal to the data similarity threshold, the reference data and the sample data are stored in the same classification data group.

[0087] Key data parameter acquisition module: after completing data storage, obtain the remaining storage space of the system; compare the remaining storage space with the remaining storage space threshold; if the remaining storage space is less than the remaining storage space threshold, analyze the historical access frequency of the data in the classified data group to obtain the key value of the data group;

[0088] Key data parameter analysis module: compares the key value of the data group with the key threshold of the data group, and determines whether the classified data group is a non-key data group based on the comparison result. If so, a compression optimization signal is generated;

[0089] Compression parameter acquisition module: based on the compression optimization signal, the non-critical data group is compressed; after the compression process, the remaining storage space is determined to be less than the remaining storage space threshold; if so, the storage duration and access times of the non-high-frequency access data in the critical data group are analyzed to obtain the compression priority value;

[0090] Compression parameter analysis module: Sort the non-high access frequency data from high to low according to the compression priority value to obtain the compression priority sequence; based on the compression priority sequence, compress the non-high access frequency data in sequence from small to large according to the serial number; when the remaining storage space reaches the remaining storage space threshold, stop compression.

[0091] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A data storage method for a memory chip, characterized in that: The following steps are involved: Step 1: Select reference data and sample data from the target storage data, analyze the data labels and data volumes of the reference data and sample data, and obtain data similarity values; The method for obtaining the data similarity value is as follows: The number of overlapping labels between each sample data and the reference data is analyzed to obtain the overlapping label ratio CH; The data volume of the reference data and the data volume of the sample data are analyzed and recorded as the data volume difference ratio CB; By analyzing the overlap label ratio CH and the data volume difference ratio CB, the formula , and obtain the data similarity value XS; where a1 and a2 are preset proportional coefficients; Step 2: Compare the data similarity value with the data similarity threshold. If the data similarity value is greater than or equal to the data similarity threshold, store the reference data and the sample data in the same classification data group; Step 3: After the data storage is completed, the remaining storage space of the system is obtained; the remaining storage space is compared with the remaining storage space threshold, and if the remaining storage space is less than the remaining storage space threshold, the historical access frequency of the data in the classified data group is analyzed to obtain the key value of the data group; The key value of the data group is obtained in the following way: The number of high-access frequency data in the classified data group is analyzed to obtain the high-access data ratio SJ; Analyze the access frequency values ​​of high access frequency data in the classified data group to obtain the high access frequency mean ratio FW; The high access data ratio SJ and the high access frequency mean ratio FW are processed and the formula is used , get the key value GJ of the data group; where b1 and b2 are preset coefficients; Step 4: Compare the data group key value with the data group key threshold, and determine whether the classified data group is a non-key data group based on the comparison result. If so, generate a compression optimization signal; Step 5: Based on the compression optimization signal, perform data compression processing on the non-critical data group; determine whether the remaining storage space after the compression processing is less than the remaining storage space threshold; if so, analyze the storage duration and access times of the non-high-access-frequency data in the critical data group to obtain a compression priority value; The compression priority value is obtained in the following manner: The non-access time of the non-high-access-frequency data in the key data group is analyzed to obtain the non-access time ratio WF; Obtain the access frequency value of the non-high access frequency data in the key data group, marked as AF; The non-visited time ratio WF and the access frequency value AF are processed and the formula is used. , get the compression priority value YX; where c1 and c2 are preset proportional coefficients; Step 6: Sort the non-high access frequency data from high to low according to the compression priority value to obtain a compression priority sequence; based on the compression priority sequence, compress the non-high access frequency data in sequence from small to large according to the serial number; when the remaining storage space reaches the remaining storage space threshold, stop compression.

2. The data storage method of a memory chip according to claim 1, characterized in that: The method for obtaining the coincidence label ratio CH is as follows: By traversing the number of sample data labels of each sample data, the number of overlapping labels of each sample data and the reference data is calculated, and the number of overlapping labels of each sample data and the reference data is obtained, which is recorded as the number of overlapping labels; The number of overlapping labels is processed by ratio with the number of reference data labels to obtain a reference overlap ratio; the number of overlapping labels is processed by ratio with the number of sample data labels of the corresponding sample data to obtain a sample overlap ratio; The reference coincidence ratio and the sample coincidence ratio are weighted summed to obtain the coincidence label ratio, which is marked as CH.

3. The data storage method of a memory chip according to claim 1, characterized in that: The data volume difference ratio CB is obtained as follows: Obtain the data volume of reference data and the data volume of sample data; The reference data volume and the sample data volume are subjected to difference processing and the absolute value is taken to obtain the data volume difference; the ratio of the data volume difference to the reference data volume is recorded as the data volume difference ratio, marked as CB.

4. The data storage method of a memory chip according to claim 1, characterized in that: The high access data ratio SJ is obtained in the following manner: Obtain the historical access times and data storage duration of the data in the classified data group; perform ratio processing on the historical access times of the data and the data storage duration to obtain the access frequency value; Compare the access frequency value with a preset access frequency threshold, and if the access frequency value is higher than the threshold, mark the data as high access frequency data; The number of high-access frequency data in the classified data group is obtained, and the number of high-access frequency data is ratio-processed with the total number of data in the classified data group to obtain a high-access data ratio, which is marked as SJ.

5. The data storage method of a memory chip according to claim 1, characterized in that: The method for obtaining the high access frequency mean ratio FW is as follows: Obtain the access frequency value of the high access frequency data in the classified data group, sum and average the access frequency values ​​of all the high access frequency data in the classified data group to obtain the high access frequency mean; perform ratio processing on the high access frequency mean and the standard high access frequency mean to obtain the high access frequency mean ratio, which is marked as FW.

6. The data storage method of a memory chip according to claim 1, characterized in that: The method for obtaining the non-visit time ratio WF is as follows: The non-access duration of the non-high-access-frequency data in the key data group is obtained, and the non-access duration is ratio-processed with the data storage duration to obtain the non-access duration ratio, which is marked as WF.

7. A data storage system of a memory chip, characterized in that: The data storage system is used to execute the data storage method described in any one of claims 1 to 6, and includes the following modules: Similarity parameter acquisition module: selects reference data and sample data from the target storage data, analyzes the data labels and data volumes of the reference data and sample data, and obtains data similarity values; Similarity parameter analysis module: compares the data similarity value with the data similarity threshold. If the data similarity value is greater than or equal to the data similarity threshold, the reference data and the sample data are stored in the same classification data group. Key data parameter acquisition module: after completing data storage, obtain the remaining storage space of the system; compare the remaining storage space with the remaining storage space threshold; if the remaining storage space is less than the remaining storage space threshold, analyze the historical access frequency of the data in the classified data group to obtain the key value of the data group; Key data parameter analysis module: compares the key value of the data group with the key threshold of the data group, and determines whether the classified data group is a non-key data group based on the comparison result. If so, a compression optimization signal is generated; Compression parameter acquisition module: based on the compression optimization signal, performs data compression processing on non-critical data groups; Determine whether the remaining storage space after compression is less than the remaining storage space threshold. If so, analyze the storage duration and access times of the non-high-access-frequency data in the key data group to obtain a compression priority value. Compression parameter analysis module: Sort the non-high access frequency data from high to low according to the compression priority value to obtain the compression priority sequence; based on the compression priority sequence, compress the non-high access frequency data in sequence from small to large according to the serial number; when the remaining storage space reaches the remaining storage space threshold, stop compression.

Images