JPH0222731A - Data compressing device - Google Patents

Abstract

PURPOSE:To prevent the increase of the data volume by comparing the number of true bytes and the number of compressed data bytes with each other to check the actual compression effect and writing true data as it is. CONSTITUTION:When receiving the indication of data transfer start from an MPU 14, a channel interface data transfer control part 13 receives data from a host CPU 11 through a channel 12 and sends data to a data compressing mechanism 18 and a channel byte counter 17 and reports the end of data transfer to the MPU 14 at the time of terminating data transfer. The mechanism 18 subjects data to run length compression or encoding compression by the indication of compression from the MPU 14 and sends the result to a device byte counter 23. Meanwhile, when receiving the report of the end of data transfer from the control part 13, the MPU 14 compares the number of true bytes from the counter 17 and the number of bytes of compressed data from the counter 23 with each other; and when the former is smaller, the MPU 14 requests the control part 13 to transfer data again and true data is stored in a data buffer 22. When the former is larger, data is compressed by the mechanism 18 and is stored in the buffer 22.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a data compression device that compresses or decompresses data.

Generally, data compression is a method for apparently increasing the storage capacity of an external storage device.

The idea behind this data compression is that, as shown in Figure 3, when data is received from the host CPU via a channel and transferred to the storage medium, the data compression mechanism (
1) run-length compression, which converts consecutively occurring characters into "character code" x n bytes of data, and (2) coded compression, which converts frequently occurring characters into shortened codes, to reduce the amount of data and reduce the amount of data on the storage medium. This increases the storage capacity of the 2nd appearance.

For example, when the true data from the host CPU is "C2C5C5C5C5C5D 1", run-length compression is performed by converting 5 bytes of "'C5": ]-code to 3 bytes of ""C5C5D1".
3FDO5'' and compresses it into data '02C5FDO5D1''.

Furthermore, as shown in Fig. 4, encoding compression converts characters that appear frequently among the original eight hit information characters, such as "space" to 3 bits, "A" to 4 bits, "Eu, i"
“D T! Compress 14M IT to 5 bits.

On the other hand, a character that appears very infrequently, for example "Z19," is increased to 13 bits (see FIG. 5).

In this case, it is necessary to consider the actual data compression effect and prevent the data amount from increasing.

[Prior Art] As a conventional magnetic tape control device having a data compression mechanism, there is one shown in FIG. 6, for example.

In FIG. 6, data from the host CPU 1 is input to a data compression mechanism 5 via a channel 2 and a channel interface 4 in a magnetic tape control device 3, and is subjected to run-length compression and/or encoded compression in the data compression mechanism 5. , the compressed data is stored in the data buffer 6. The compressed data is then written onto the magnetic tape 8 by the read/write controller 1m7.

The compressed data written on the magnetic tape 8 is stored in the data buffer 76 by the read/write control mechanism 7, and then restored to true data by the data restoration mechanism 9 and sent to the host via the channel interface 4 and channel 2. C to l
P (read out in J 1.

[Problems to be Solved by the Invention] However, in such conventional magnetic tape control devices, when compressing data, there is a problem of incompatibility between the compression algorithm and the nature of the data, which causes the compression effect to be ineffective or not. teeth,
When looking at one piece of data, it does not change dynamically,
Rather, it is seen as a data characteristic that appears throughout the data and is expected to persist until the end of the data.

For this reason, when data compression is performed unilaterally, there is a problem in that the amount of data increases due to the incompatibility between the compression algorithm and the nature of the data.

The present invention has been made in view of such conventional problems, and is a data system that prevents an increase in the amount of data by importing and outputting true data as is when the amount of data increases. The present invention aims to provide a pressure fi-i device.

[Means to solve the problem] FIG. 1 is a basic configuration diagram of the present invention.

In FIG. 1, 18 is a data compression means for compressing data from the host CPU 11, 22 is a holding means for holding compressed data or uncompressed data, and 17 is a host CPU
first counting means for counting the number of bytes of data from I; 2;
3 is a second counting means for counting the number of bytes of compressed dredging data, 14A is a comparison means for comparing the output of the first counting means 17 and the output of the second counting means 23, and 14B is a comparison means of the comparison means 14A. An instructing means 14C instructs the data compressing means 18 to compress or non-compress data depending on the result, when the output of the first counting means 17 is larger than the output of the second counting means 23, the instruction means 14C instructs the host CPU 11 to re-compress the data. This is a request means for requesting transfer.

[Operation] In the present invention, the number of bytes of true data and the number of bytes after compression are compared, and if the number of bytes of true data is larger than the number of bytes of data after compression, The data size increases due to the incompatibility between the compression algorithm and the nature of the data. In some cases, it is possible to prevent an increase in the amount of data.

That is, since the actual compression effect is checked and data is compressed or uncompressed, an increase in the amount of data can be prevented.

[Example] Embodiments of the present invention will be described below based on the drawings.

FIG. 2 is a block diagram showing one embodiment of the present invention.

First, to explain the configuration, in FIG. 2, 11 is a host CPU, 12 is a channel, and 13 is a channel interface data transfer control section (hereinafter referred to as data transfer control section). The data transfer control unit 13 controls data transfer with the host CPU 11 via the channel 12.

The data transfer control unit 13 is a microprocessor (hereinafter referred to as M
When receiving an instruction to start data transfer from the CPU (PU) 14 via the signal line 15, the data is received from the host CPU 11 via the channel 12 and transferred via the bus 16 to the channel byte counter (first counting means) 17 and The data is sent to the data compression mechanism (data compression means) 18.

Further, when the data transfer is completed, the data transfer control unit 13 instructs the MPU 1 to terminate the data transfer via the signal line 19.
Notify 4. Further, the data transfer control unit 13
When receiving an instruction to retransfer data from 14 via signal line 20, the data is retransferred to channel byte counter 17 and data compression R mechanism 18.

The digital byte counter 17 counts the number of bytes of data (true data) sent from the data transfer control section 13 and outputs the number of true data to the MPU 14.

The data compression mechanism 18 receives an instruction to compress or uncompress data from the MPU 14 via a signal line 21, compresses the data by run-length compression and/or coded compression according to the compression instruction, and stores the compressed data in a data buffer (holding 22 and a device byte counter (second counting means) 23, and in response to a non-compression instruction, the true data is output as is to the data buffer 22.

Compressed data or uncompressed data stored in the data buffer 22 is read onto a magnetic tape (not shown) by a read/write control mechanism (not shown).

The device byte counter 23 counts the number of bytes of compressed data output from the data compression mechanism 18 and outputs the number of bytes to the MPU 14.

Upon receiving the data transfer end notification from the data transfer control unit 13, the MPU 14 calculates the true number of data bytes from the channel byte counter 17 and the device byte counter 2.
Compare the number of bytes of compressed data from 3 to check the compression effect.

The MP'U 14 determines that the number of bytes in the channel byte counter 17 ≦ the number of bytes in the device byte counter 23 .
Instructs data compression mechanism 18 to compress, and registers channel byte counter 17 pie 1 to number > device byte counter 23
When the number of bytes is , it requests the data transfer control unit 13 to retransfer the data and instructs the data compression mechanism 18 to decompress, allowing the true data to pass through as is.

Therefore, the MPU 14 inputs counter 1 to channel by 1.
Comparison means 14A compares the output of device byte counter 23 with the output of device byte counter 23, and data compression mechanism 18
It functions as an instructing means 14B for instructing compression or non-compression, and a requesting means 14C for requesting the data transfer control section 13 to retransmit data.

Next, the operation will be explained.

First, upon receiving an instruction to start data transfer from the MPU 14, the data transfer control unit 13 receives data from the host CPU 11 via the channel 12 and sends the data to the data compression mechanism 1.
8 and channel byte counter 17, and when the data transfer is completed, the MPU 14 is notified of this.

The data compression mechanism 18 performs run-length compression and/or code compression on data according to a compression instruction from the MPU 14 and sends the compressed data to the data buffer 22 and device byte counter 23 .

On the other hand, upon receiving the data transfer end notification from the data transfer control unit 13, the MPU 14 compares the number of bytes of true data from channel byte 1 to counter 17 with the number of bytes of compressed data from device byte counter 23. However, when the number of bytes in the channel byte counter 17 is equal to the number of bytes in the device byte counter 23, the data transfer control unit 1
Data compression mechanism 1 requests data retransfer to data compression mechanism 3.
B is instructed to uncompress the data, and the true data is passed through as is and stored in the data buffer 22.

Note that when the number of bytes in the channel byte counter 17 > the number of bytes in the device byte counter 23, the MPU 14
instructs the data compression mechanism 18 to compress and stores the compressed data in the data buffer 22.

Therefore, if the amount of data increases due to an incompatibility between the compression algorithm and the nature of the data, it is possible to prevent the amount of data from increasing by writing the true data as is from the data buffer 122 to the magnetic tape.

Note that when the number of bytes in the channel byte counter 17≧the number of bytes in the device byte counter 23, the amount of data can be reduced by performing data compression.

[Effects of the Invention] As explained above, according to the present invention, the true number of data bytes and the number of compressed data bytes are compared to check the actual compression effect, so that when the amount of data increases, The true data can be written as is, and an increase in the amount of data can be prevented.

[Brief explanation of the drawing]

Figure 1 is a basic configuration diagram of the present invention, Figure 2 is a block diagram showing an embodiment of the present invention, Figure 3 is an explanatory diagram explaining the concept of data compression, and Figure 4 is the concept of coding compression. FIG. 5 is a graph explaining the concept of coding compression, and FIG. 6 is a block diagram showing a conventional example. In the figure, 11: host CPU. 12: Channel, 13: Data transfer control section, 14.14A to 14C: Microprocessor (comparison means, instruction means, request means), 15: Signal line, 16: Bus, 17: Channel byte counter (first counting means) , 18
: data compression mechanism (data compression means), 19-21: signal line, 22: data buffer 1 (holding means), 23: device byte counter (second counting means). To the side of the steamer (2) data hill m'hJjJε 兇-月froJ ひろ月 11W fig. Sentence - Mouth Moon ■ Ukun ← Moon Figure 4 Coding r' Box How to Serve ε Theory Graph Figure 5 I Bi Old Book U & Water Sen 70 Tsuku Times Figure 6

Claims (1)

[Claims] data compression means (18) for compressing data from the host CPU (11); holding means (22) for holding compressed data or uncompressed data; and a data compression means (22) for counting the number of bytes of data from the host CPU (11). 1 counting means (17), and a second counting means (17) for counting the number of bytes of data after compression.
23), the output of the first counting means (17) and the second
Comparison means (14A) for comparing the output of the counting means (23)
and instruction means (14B) for instructing the data compression means (18) to compress or uncompress data based on the comparison result of the comparison means (14A), and the output of the first counting means (17) is calculated by a second counting means. A data compression device comprising request means (14C) for requesting a host CPU (11) to retransmit data when the output is larger than the output of the means (23).