JPH04256210A - Data compression/storage system - Google Patents

Abstract

PURPOSE:To supply a data compression/storage system which records a difference in a storage device and which can fetch it at high speed. CONSTITUTION:A depending relation table 2 expressed by hierarchy structure in which complete data is set to be a root node and the difference of other data to be a node and which registers a depending relation as to difference data of respective nodes is provided. Data on the node in the lowest layer of hierarchy structure 2 in which complete data is set to be the root node and the difference of the other data is set to be the node at the time of storing data is recorded in the storage device, the depending relation table 2 is referred to at the time of reading, all data being an object and data of the node depending on said objective data are fetched and they are restored to original data.

Description

[Detailed description of the invention]

0001

[Field of Industrial Application] The present invention relates to a data compression and storage method for storing and compressing data differences, and more particularly, relates to a data compression and storage method for compressing information (data) by taking the differences between pieces of information. be. When storing information (data) based on differences between pieces of information in secondary memory, and when accessing intermediate information, the information must be read sequentially even if the secondary memory has a random access function. First, it becomes slow to retrieve the desired information. For this reason, it is desired to use some method to retrieve target information at a speed close to random access.

0002

2. Description of the Related Art Conventional data compression and storage systems attempt to compress data by storing complete information at the beginning or end and storing differential information at other locations. However, with this method, when extracting information in the middle, after reading the first or last complete information, sequentially reading differential information one after another to create complete information, and finally generating the target information. I was trying to do that.

0003

[Problems to be Solved by the Invention] Therefore, a lot of time is required to read the difference information, and there is a problem in that the target information can only be read out at a low speed. As a result, even if the secondary storage device is capable of random access, this advantage cannot be taken advantage of, and it takes time to retrieve the desired data.

SUMMARY OF THE INVENTION An object of the present invention is to provide a data compression storage method that allows differences to be recorded in a storage device and retrieved at high speed.

[0005]

[Means for Solving the Problems] Means for solving the problems will be explained with reference to FIGS. 1 and 3. In FIG. 1, hierarchical structure 1 is a structure in which complete data is the root node and each node has its differential data, and is a structure for storing the data of the lowest layer node in a storage device. .

[0006] The dependency relationship table 2 is a table in which dependency relationships of differential data of nodes at the lowest layer of the hierarchical structure 1 are registered. In FIG. 3, complete data 3 is complete data inserted into the differential data.

[0007]

[Operation] As shown in FIG. 1, when storing data, the present invention expresses complete data in a hierarchical structure 1 with the root node and differences in other data as nodes, and stores the data in the lowest layer node. At the same time as recording the difference data in the device, the dependency relationships of the recorded difference data are registered in the dependency relationship table 2, and when reading data, this dependency relationship table 2 is referred to and all data with dependency relationships are extracted and these are also I am trying to restore the original data. Also,
As shown in FIG. 3, starting from complete data 3 and sequentially recording the difference data in the storage device is repeatedly performed, and when reading, the difference data sequentially read from the neighboring complete data 3 is also recorded. I am trying to restore the original data.

Therefore, it is possible to record the difference data in a storage device and restore the original data by extracting the dependent data based on the dependency relationship table 1, or to restore the original data sequentially based on the complete data 3 in the vicinity. By extracting the difference data and restoring the original data, it becomes possible to extract the data at high speed and restore the original data.

[0009]

Embodiment Next, the configuration and operation of an embodiment of the present invention will be explained in detail using FIGS. 1 and 2. Figure 1 (a)
shows an example of hierarchical structure 1. In (a) of FIG. 1, the hierarchical structure 1 has complete data as the root node, each node has its differential data as shown in the figure, and the data of the lowest layer node is recorded in the storage device, The original data can be restored from these recorded data.

FIG. 1B shows an example of the dependency relationship table 2. This dependency relationship table 2 is a table for registering on which data the difference data of the lowest layer node of the hierarchical structure 1 in FIG. 1(A) depends. For example, as shown in the figure, the data number "4" is the dependent number "3".
”, and this dependent data number “3” becomes the dependent number “1”, and since there is no number that depends on this dependent data number “1”, as a result, the data number “4”
is created from the difference between two data numbers “3” and “1”, so by reversing this operation, the original data number “4” is created.
” can be restored. Specifically, data number “1” can be restored.
”, data number “3” is added to the complete data of the storage device location.
The original data can be restored by adding the data at the location of the storage device with data number "4" and then adding the data at the location of the storage device with data number "4". This is the location (address) where the

FIG. 1C shows an example of the data format. Here, the data is composed of a data number, a dependent number, and differential data, and corresponds to the data number, dependent number, and differential data stored in the data position in (b) of FIG. Next, the operation of the configuration of FIG. 1 will be explained in detail in accordance with the order shown in the flowchart of FIG. 2(a). Here, (b) in FIG. 2 is data number 4 in (b) of FIG.
This figure specifically shows how to restore the data.

In FIG. 2A, S1 is assumed to be i=target number. For example, in (b) of Figure 2 on the right side, i=
As shown as 4, the data number “4” in (a) of FIG.
Let be the target number of data recovery. S2 determines whether there is a dependent number. If YES, put the number i into the stack 4 in S3, set i=dependent number in S4, repeat S2, and put all the dependent numbers into the stack 4. On the other hand, if there are no more dependent numbers and the answer is NO, the process advances to S5 and repeats S6 to S8 to restore the data.

[0013] To specifically explain S1 to S4, as shown in the upper part of (b) of FIG.
Based on = 4, we found the dependent number "3" by referring to (b) dependency table 2 in Figure 1, so we put this 4 in stack 4 and set i = 3 (YES in S2, S3
, S4).

Second time: Based on i=3, we looked up the dependency table 3 in Figure 1 (b) and found the dependent number "1", so we put this 3 in the stack 4 and set i=1. (YES in S2, S3, S4). 3rd time: i=1
Since there is no dependent number based on (b) dependency table 3 in FIG. 1 (NO in S2), the process proceeds to S5 to S8 and data restoration is started. In this state,
In the stack 4, 4 and 3 are placed as shown.

[0015] S5 reads number 1. This reads the value of the first data (complete data) from the storage device. S6 determines whether the stack 4 is empty. If NO, the data at the head of stack 4 (here, for example, data number 3 from the storage device) is read in in S7, and the data is restored (for example, the data numbered 1 and the data numbered 3 are added together). (restore the original number 3 data), pop the stack in S8 and remove the first value, repeat S6 and subsequent steps, read the data of the number sequentially taken out from stack 4 from the storage device, and restore the differential data. I do. Then, when stack 4 becomes empty and S6 becomes YES,
Finish the restoration.

To explain S5 to S8 in detail, as shown in the lower part of FIG. Add the complete data and the differential data of 3 to restore the data of 3), pop the stack 4 and remove 3.

Second time: Read data 4 at the top of the stack from the storage device and restore 4 (add the restored data 3 and the differential data 4 to restore data 4),
Pop stack 4 and remove 4. Third time: Since the stack is empty and the answer to S6 is NO, the target data of 4 has been restored at this point.

As described above, the target number (for example, i
= 4), refer to (b) dependency table 2 in Figure 1 and put all the dependent numbers into stack 4 until complete data is reached (S1 to S4), and then It becomes possible to restore data of a target number by sequentially restoring complete data based on the number.

Next, the structure and operation of another embodiment of the present invention will be explained in detail using FIG. 3. FIG. 3(A) shows an example of the data structure. Here, complete data 3 is
This is data itself that is not differential data. By inserting complete data at various locations in this manner, it becomes possible to sequentially read the differential data starting from the closest complete data 3 with the target number i and restore the original data.

FIG. 3(b) schematically shows how, when a plurality of complete data 3 of FIG. 3(a) are provided, these complete data 3 are inserted at various locations as shown in the figure. . By recording the complete data 3 in various places in the storage device in this way, it is read out sequentially from the position (address) of the complete data 3 closest to the target difference data number, and for example, the value is added sequentially. can be restored.

[0021]

[Effects of the Invention] As explained above, according to the present invention,
Record the differential data in a storage device, and restore the original data by extracting the dependent data based on the dependency table 2, or sequentially extract the differential data based on the neighboring complete data 3. Since the system employs a configuration in which the original data can be restored using the system, data can be retrieved at high speed and the original data can be quickly restored. This eliminates the need to read the differential data in a completely sequential manner to restore the original data as in the past, and because the data is imported randomly for many parts, data retrieval is dramatically faster, and as a result, It becomes possible to speed up data reading.

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram of one embodiment of the present invention.

FIG. 2 is an explanatory diagram of the configuration of one embodiment of the present invention.

FIG. 3 is a configuration diagram of another embodiment of the present invention.

[Explanation of symbols]

1: Hierarchical structure 2: Dependency table 3: Complete data 4: Stack

Claims (2)

[Claims] Claim 1: In a data compression storage method that stores and compresses differences in data, a hierarchical structure (1) in which complete data is the root node and differences in other data are nodes.
A dependency table (2) is provided to register the dependencies of the difference data of each node, and when data is stored, the above hierarchical structure (2) is created in which the complete data is the root node and the differences of other data are nodes. ) is recorded in the storage device, and when reading it, the above dependency table (2) is referred to, the target data and all the data of the nodes that depend on it are retrieved, and the data is restored to the original data. A data compression storage method characterized by being configured as follows. Claim 2: In a data compression storage method that stores and compresses data differences, starting from complete data (3), the difference data is repeatedly recorded in a storage device sequentially, and when read out, A data compression storage method characterized in that the original data is restored based on differential data sequentially read from neighboring complete data (3).