JP2023128061A - Information processing apparatus, information processing method, and information processing program - Google Patents

Abstract

To provide an information processing apparatus, an information processing method, and an information processing program that can avoid considerable extension of alignment time when determining a reading order considering the priority of reading.SOLUTION: When a processor derives a reading order for a plurality of pieces of data to be read which is recorded in a magnetic tape and for which priority of reading is set by using a predetermined algorithm, the processor performs, based on the priority, weighting on an index value according to recording positions of the pieces of data to be read, the value being used in determining the reading order by using the algorithm, and derives the reading order of the plurality of pieces of data to be read based on the weighted index value.SELECTED DRAWING: Figure 4

Description

The disclosed technology relates to an information processing device, an information processing method, and an information processing program.

The following techniques are known as techniques related to processing for determining the reading order of data recorded on a recording medium. For example, Patent Document 1 discloses a file restoration order rearranging unit that acquires the recording position and recording direction of a file to be read, and determines the read order of the file based on the acquired recording position and recording direction; A tape control device is described that includes a file restoration section that reproduces files to be read in the order rearranged by the restoration sorting section. The file restoration order rearranging unit determines the reading order of files based on the recording position and recording direction of the files so that the seek time of the file to be read is the shortest.

Patent Document 2 discloses that when a read request including first priority information is processed, the read request including the second priority information is processed within the delay limit value of the already arrived read request including the second priority information. A storage device control device is described that outputs a read command with priority over a read request including a read command.

Japanese Patent Application Publication No. 2012-9105 Japanese Patent Application Publication No. 10-275059

The RAO (Recommended Access Order) function is known as a method for shortening the time required for positioning the head of a tape drive (hereinafter referred to as positioning time) when reading data recorded on a magnetic tape. The RAO function is a function that determines the read order of a plurality of pieces of data recorded on a magnetic tape to be read using a predetermined order derivation algorithm so that alignment time is shortened.

A reading priority may be set for the data to be read, and data with a high reading priority is requested to be read preferentially from the magnetic tape. However, existing order derivation algorithms do not take into account the priority set for the data to be read. On the other hand, if the read order of data with a high priority is simply set higher, the alignment time may become significantly longer.

The disclosed technique has been developed in view of the above points, and aims to avoid a significant increase in alignment time when determining the readout order in consideration of the readout priority.

An information processing device according to the disclosed technology includes at least one processor. When a processor uses a predetermined algorithm to derive a read order for multiple pieces of data to be read that are recorded on a magnetic tape and have read priorities set, the processor determines the read order using an algorithm. The index values corresponding to the recording position of each data to be read to be used are weighted based on the priority, and the reading order of the plurality of data to be read is derived based on the weighted index values.

The index value may be a value that correlates with the moving time or moving distance when the head moves to the recording position of the data. The processor may weight data with relatively high priority using a smaller weighting coefficient than data with relatively low priority.

The information processing method according to the disclosed technology uses an algorithm when deriving a read order using a predetermined algorithm for a plurality of pieces of data to be read that are recorded on a magnetic tape and have read priorities set. Index values corresponding to the recording positions of each data to be read used when determining the read order are weighted based on the priority, and the read order of multiple data to be read is determined based on the weighted index values. At least one processor included in the information processing device executes the deriving process.

The information processing program according to the disclosed technology uses an algorithm when deriving a read order using a predetermined algorithm for a plurality of pieces of data to be read that are recorded on a magnetic tape and have read priorities set. Index values corresponding to the recording positions of each data to be read used when determining the read order are weighted based on the priority, and the read order of multiple data to be read is determined based on the weighted index values. This is a program for causing at least one processor included in an information processing device to execute a process to be derived.

According to the disclosed technology, when determining the readout order in consideration of the readout priority, it is possible to avoid a significant increase in alignment time.

1 is a diagram illustrating an example of a configuration of a storage system according to an embodiment of the disclosed technology. FIG. 3 is a diagram illustrating an example of the content of management information according to an embodiment of the disclosed technology. FIG. 1 is a diagram illustrating an example of a hardware configuration of a tape drive according to an embodiment of the disclosed technology. FIG. 2 is a functional block diagram illustrating an example of a functional configuration of a tape drive according to an embodiment of the disclosed technology. FIG. 3 is a diagram illustrating an example of weighting of index values according to an embodiment of the disclosed technology. 2 is a flowchart illustrating an example of the flow of order derivation processing according to an embodiment of the disclosed technology. FIG. 7 is a diagram illustrating an example of a moving path of a magnetic head when data to be read is read in accordance with a read order derived without considering the read priority. FIG. 3 is a diagram illustrating an example of a moving path of a magnetic head when data to be read is read in accordance with a read order derived by taking read priorities into consideration.

An example of an embodiment of the disclosed technology will be described below with reference to the drawings. In addition, the same reference numerals are given to the same or equivalent components and parts in each drawing, and overlapping explanation will be omitted.

FIG. 1 is a diagram showing an example of the configuration of a storage system 1 according to an embodiment of the disclosed technology. The storage system 1 stores data that is requested to be saved from an external device (not shown) such as a user terminal connected to the storage system 1 via a network, and when there is a request to read data from the external device, Read the requested data and send it to the external device. The storage system 1 may be an object storage system that handles data in units of objects. An object is configured to include a data body and metadata regarding the data body.

Storage system 1 includes a server 10 and a tape library 20. The tape library 20 includes a plurality of slots (not shown) and a plurality of tape drives 40, and a magnetic tape 30 is stored in each slot. Data requested to be stored from an external device such as a user terminal is recorded on the magnetic tape 30. The tape drive 40 records (writes) data on the magnetic tape 30 loaded in the tape drive 40 and reads data from the magnetic tape 30 based on instructions from the server 10 . An example of the magnetic tape 30 is an LTO (Linear Tape-Open) tape. When recording or reading data on the magnetic tape 30, the target magnetic tape 30 is taken out of the slot and loaded into a predetermined tape drive 40. When the recording or reading of data on the magnetic tape 30 loaded in the tape drive 40 is completed, the magnetic tape 30 is taken out from the tape drive 40 and stored in a predetermined slot. The server 10 controls recording and reading of data on the magnetic tape 30.

The server 10 accepts data read requests from external devices (not shown) such as user terminals. A read deadline may be set for the data to be read. The read deadline is set for each piece of data by, for example, a user who issues a read request, and is recorded in a nonvolatile memory (not shown) included in the server 10. The server 10 sets a read priority for the data to be read related to the read request. The server 10 sets priorities for each piece of data to be read according to the read deadline set for each piece of data. The server 10 generates a data list in which data ID (identification), read deadline, and priority are associated with each other for each piece of data to be read, and stores this in a non-volatile memory (not shown) provided in the server 10. do.

FIG. 2 is a diagram showing an example of the contents of the above data list. The data list 120 is information in which a data ID for identifying the data, a read deadline set for the data, and a read priority set according to the read deadline are associated with each other for the read target data for which a read request has been made. be. The read deadline is set for each data, for example, by a user who issues a read request. The priority is set according to the period from the current time to the reading deadline. In this embodiment, priority 1 is a group with the shortest period until the read deadline, and data to which priority 1 is set has the highest degree of read priority. Priority 3 is a group with the longest period until the read deadline, and data set with priority 3 has the lowest read priority. Although FIG. 2 shows an example where the data to be read is classified into three priority groups according to the read deadline, the data may be classified into two or four or more priority groups. .

The server 10 issues an instruction to the tape drive 40 to derive a read order (hereinafter referred to as an order derivation instruction) using the RAO function provided in the tape drive 40 for a plurality of pieces of data to be read related to the received read request. The server 10 sends the data list 120 to the tape drive 40 along with the order derivation instruction.

FIG. 3 is a diagram showing an example of the hardware configuration of the tape drive 40. The tape drive 40 records (writes) data on the magnetic tape 30 and reads data from the magnetic tape 30 based on instructions from the server 10 . A magnetic tape cartridge 31 containing a magnetic tape 30 is loaded into the tape drive 40 . The tape drive 40 includes a magnetic head 41 that performs recording and reproduction on the magnetic tape 30, and a control section 42 that performs recording and reproduction control on the magnetic tape 30. The control unit 42 includes a processor 43 such as a PLD (Programmable Logic Device), and a nonvolatile memory 45 that stores an information processing program 44 as firmware. The tape drive 40 is an example of an information processing device in the disclosed technology.

The tape drive 40 has an RAO function to reduce the time required for positioning the magnetic head 41 (positioning time) when reading data recorded on the magnetic tape 30. That is, when the tape drive 40 receives an instruction to read a plurality of data recorded on the magnetic tape 30, the tape drive 40 uses a nearest neighbor method, a pairwise method, etc. to shorten the positioning time for the plurality of data to be read. A read order is derived using a known order derivation algorithm. Note that the nearest neighbor method is a method of sequentially reading a plurality of pieces of data to be read starting from the data located closest to the magnetic head. The pairwise exchange method is a method of evaluating the read time by changing the read order of the data to be read last in the read order derived by the nearest neighbor method to the order of the data to be read at the beginning or among other consecutive data. In the pairwise exchange method, the readout order that results in the shortest evaluated readout time is derived as the optimal readout order. Note that the nearest neighbor method and the pairwise exchange method are also described in JP-A-2012-128937.

FIG. 4 is a functional block diagram showing an example of the functional configuration of the tape drive 40 when reading data to be read from the magnetic tape 30. The tape drive 40 includes an acquisition section 401, an order derivation section 402, and a reading section 403. When the processor 43 executes the information processing program 44, the tape drive 40 functions as an acquisition section 401, an order derivation section 402, and a reading section 403.

The acquisition unit 401 acquires the data list 120 supplied from the server 10.

The order deriving unit 402 derives a read order for a plurality of pieces of data to be read using a predetermined order deriving algorithm in response to an order deriving instruction from the server 10 . The order deriving unit 402 performs the following processing to derive a read order for a plurality of data to be read. In the following description, a case will be exemplified in which the nearest neighbor method is used as the order derivation algorithm.

The order derivation unit 402 derives, for each data to be read, an index value according to the recording position of the data. The recording position of the data to be read can be specified from the data ID recorded in the data list 120. The above index value is, for example, a value that correlates with the movement time (seek time) or the movement distance (seek distance) when the magnetic head 41 moves from the current position to the data recording position. The movement time (seek time) and the movement distance (seek distance) can be estimated from the recording position of each data by the RAO function provided in the tape drive 40. FIG. 5 illustrates moving distances as index values derived for each of data #1 to #10 to be read. The moving distance may be an amount expressed by the number of data blocks partitioned by a predetermined data size.

The order derivation unit 402 weights the index values derived for each data to be read based on the priority set for the data. Weighting is performed using a weighting coefficient determined according to the priority set for each data to be read. The order deriving unit 402 weights data with relatively high priority using a smaller weighting coefficient than data with relatively low priority. Figure 5 shows a case where a weighting factor of 0.5 is applied to priority 1, a weighting factor of 0.7 is applied to priority 2, and a weighting factor of 1 is applied to priority 3. Illustrated. The priority set for each data to be read can be specified from the data list 120.

The order derivation unit 402 derives a weighted index value by multiplying the index value derived for each data to be read by the weighting coefficient applied to the data. In FIG. 5, a correction value obtained by multiplying the travel distance derived for each data by the corresponding weighting coefficient is illustrated as a weighted index value.

The order deriving unit 402 determines the data with the smallest weighted index value as the data to be read first. In the example shown in FIG. 5, data #4 having the smallest correction value is determined as the data to be read first.

The order derivation unit 402 assumes that the recording position of the data determined as the first data to be read is the current position of the magnetic head 41, derives the index value in the same manner as described above, and calculates the derived index value based on the priority. Weighting is performed, and data to be read out second is determined based on the weighted index value. The order deriving unit 402 determines the read order for all the data to be read by repeating the above process. The order deriving unit 402 transmits information indicating the derived reading order of all data to the server 10.

The reading unit 403 reads data to be read from the magnetic tape 30 in accordance with the reading order derived by the order deriving unit 402 in response to a read instruction issued from the server 10 . The reading unit 403 transmits the data read from the magnetic tape 30 to the server 10.

FIG. 6 is a flowchart illustrating an example of the sequence derivation process performed by the processor 43 of the tape drive 40 executing the information processing program 44. The information processing program 44 is executed, for example, when an order derivation instruction is issued from the server 10.

In step S1, the acquisition unit 401 acquires the data list 120 supplied from the server 10 together with the order derivation instruction.

In step S2, the order deriving unit 402 assigns an initial value of 1 to a variable i that specifies the data reading order. As a result, a process is started to determine which data is read first among the data to be read.

In step S3, the order derivation unit 402 derives, for each data to be read, an index value according to the recording position of the data. The recording position of the data to be read can be specified from the data ID recorded in the data list 120. The index value is, for example, a value that correlates with the movement time (seek time) or movement distance (seek distance) when the magnetic head 41 moves from the current position to the data recording position.

In step S4, the order derivation unit 402 weights the index value for each data derived in step S3 based on the priority set for the data. Weighting is performed using a weighting coefficient determined according to the priority set for each data to be read. The order deriving unit 402 weights data with relatively high priority using a smaller weighting coefficient than data with relatively low priority. The order derivation unit 402 derives a weighted index value by multiplying the index value derived for each data to be read by the weighting coefficient applied to the data.

In step S5, the order deriving unit 402 determines the data whose reading order is i-th among the plurality of data to be read based on the weighted index value. Here, the data with the smallest weighted index value is determined as the data to be read first.

In step S6, the order deriving unit 402 determines whether the read order has been determined for all of the data to be read. If it is determined that the read order has not been determined for all of the data to be read, the process moves to step S7, and if it is determined that the read order has been determined for all of the data to be read, the process moves to step S8. will be moved to

In step S7, the order deriving unit 402 increments a variable i that specifies the data reading order, and returns the process to step S3. This starts the process of determining the second data to be read out. The processes from step S3 to step S7 are repeated until the read order is determined for all data to be read.

In step S8, the order deriving unit 402 transmits information indicating the read order derived for all data to be read to the server 10. Having received this information, the server 10 issues an instruction to the tape drive 40 to read the data to be read.

In step S9, the reading unit 403 determines whether a reading instruction has been issued from the server 10. If it is determined that a read instruction has been issued from the server 10, the process moves to step S10.

In step S10, the reading unit 403 reads data to be read from the magnetic tape 30 in accordance with the reading order derived through the repeated processing from step S3 to step S7, and transmits the read data to the server 10.

FIG. 7A is a diagram illustrating an example of a moving path of the magnetic head 41 when data #1 to #6 to be read are read in accordance with the read order derived without considering the read priority. FIG. 7B is a diagram illustrating an example of a moving path of the magnetic head 41 when data #1 to #6 to be read are read in accordance with a read order derived in consideration of read priority. 7A and 7B each illustrate a case where the nearest neighbor method is used as the order derivation algorithm. Also shown are values indicating the read order of data #1 to #6 to be read.

If the priority is not taken into account, the data closest to the current position of the magnetic head 41 is sequentially read. That is, as shown in FIG. 7A, data #1, #2, #3, #4, #5, and #6 are read out in this order. On the other hand, when taking priority into consideration, data with the smallest index value weighted based on the priority is sequentially read out. FIG. 7B illustrates a case where data #5 is set first in the read order as a result of weighting index values correlated with the travel time or travel distance of the magnetic head 41 based on priority.

As described above, the processor 43 included in the tape drive 40 according to the embodiment of the disclosed technology uses a predetermined algorithm to read a plurality of pieces of data to be read that are recorded on the magnetic tape 30 and have priorities set. When deriving the order, the index values according to the recording position of each data to be read used when determining the read order using an algorithm are weighted based on the priority, and the index values are calculated based on the weighted index values. Then, the reading order of the plurality of data to be read is derived. The index value is a value that correlates with the moving time or moving distance when the magnetic head 41 moves to the recording position of the data. The processor 43 weights data with relatively high priority using a smaller weighting coefficient than data with relatively low priority.

According to the tape drive 40 according to the present embodiment, reading of a plurality of data to be read is performed based on the index value weighted according to the priority for use in determining the read order using an algorithm. Since the order is derived, it is possible to avoid a significant increase in alignment time when determining the readout order in consideration of the readout priority.

Note that in this embodiment, a case has been exemplified in which the tape drive 40 derives the read order using the order derivation algorithm, but the server 10 may also perform this. Further, in this embodiment, the case where the nearest neighbor method is used as the order derivation algorithm is exemplified, but it is also possible to use other algorithms.

In each of the above embodiments, the hardware structure of a processing unit that executes various processes, such as the acquisition unit 401, order derivation unit 402, and readout unit 403, includes the following various processors ( processor) can be used. As mentioned above, the various processors mentioned above include the CPU, which is a general-purpose processor that executes software (programs) and functions as various processing units, as well as processors such as FPGAs whose circuit configuration can be changed after manufacturing. These include programmable logic devices (PLDs), ASICs (Application Specific Integrated Circuits), and other specialized electrical circuits that are processors with circuit configurations specifically designed to execute specific processes.

One processing unit may be composed of one of these various processors, or a combination of two or more processors of the same type or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). combination). Further, the plurality of processing units may be configured with one processor.

As an example of configuring multiple processing units with one processor, firstly, one processor is configured with a combination of one or more CPUs and software, as typified by computers such as a client and a server. There is a form in which a processor functions as multiple processing units. Second, there are processors that use a single IC (Integrated Circuit) chip, such as System on Chip (SoC), which implements the functions of an entire system including multiple processing units. be. In this way, various processing units are configured using one or more of the various processors described above as a hardware structure.

Furthermore, as the hardware structure of these various processors, more specifically, an electric circuit (circuitry) that is a combination of circuit elements such as semiconductor elements can be used.

Further, in the embodiment described above, a mode has been described in which the information processing program 44 is stored (installed) in the memory 45 in advance, but the present invention is not limited to this. The information processing program 44 is provided in a form recorded on a recording medium such as a CD-ROM (Compact Disc Read Only Memory), a DVD-ROM (Digital Versatile Disc Read Only Memory), or a USB (Universal Serial Bus) memory. Good too. Further, the information processing program 44 may be downloaded from an external device via a network.

1 Storage system 10 Server 20 Tape library 30 Magnetic tape 31 Magnetic tape cartridge 40 Tape drive 41 Magnetic head 42 Control unit 43 Processor 44 Information processing program 45 Memory 120 Data list 401 Acquisition unit 402 Sequence derivation unit 403 Reading unit

Claims (5)

An information processing device comprising at least one processor,
When the processor derives a read order using a predetermined algorithm for a plurality of pieces of data to be read that are recorded on a magnetic tape and have read priorities set,
weighting index values according to the recording position of each data to be read used when determining the read order using the algorithm based on the priority;
An information processing device that derives a reading order of a plurality of data to be read based on the weighted index value. The information processing apparatus according to claim 1, wherein the index value is a value that correlates with a moving time or a moving distance when the head moves to the recording position of the data. The information processing apparatus according to claim 1 or 2, wherein the processor performs the weighting on the data having the relatively high priority using a smaller weighting coefficient than the data having the relatively low priority. When deriving the read order using a predetermined algorithm for multiple pieces of data to be read that are recorded on magnetic tape and have read priorities set,
weighting index values according to the recording position of each data to be read, which are used when determining the read order using the algorithm, based on the priority;
An information processing method in which at least one processor included in an information processing apparatus executes a process of deriving a reading order of a plurality of data to be read based on the weighted index value. When deriving the read order using a predetermined algorithm for multiple pieces of data to be read that are recorded on magnetic tape and have read priorities set,
weighting index values according to the recording position of each data to be read, which are used when determining the read order using the algorithm, based on the priority;
An information processing program for causing at least one processor included in an information processing apparatus to execute a process of deriving a reading order of a plurality of data to be read based on the weighted index value.

Images