JP6848741B2 - Storage test equipment and storage test program - Google Patents

Description

The present invention relates to a storage test apparatus and a storage test program.

In recent years, the amount of data stored in a storage device has increased due to the use of big data and the like. In order to cope with the increasing amount of data, the storage device reduces the amount of data stored by using a data deduplication function for removing data duplication and a data compression function for compressing data. Further, in order to reduce the initial investment, the storage device is provided with a capacity virtualization function (Thin Provisioning) that allocates physical storage only to the capacity to be used for the virtually determined total capacity.

FIG. 12 is a diagram for explaining the reduction of the data storage amount. In FIG. 12, the information processing device is a computer that stores and uses data as a logical volume in a storage device. As shown in FIG. 12, since the data that the information processing apparatus writes for the first time includes the data pattern # 1, the data pattern # 2, and the data pattern # 1, the second data pattern # 1 that is duplicated by the deduplication function Is deduplicated.

Then, the data pattern # 1 and the data pattern # 2 are each compressed by the data compression function and written to the disk 9. At this time, a chunk-sized disk area is allocated by the capacity virtualization function, and the compressed data pattern # 1 and data pattern # 2 are written to the allocated disk area. The chunk size is an allocation unit allocated from the disk 9 by the capacity virtualization function. The chunk-sized disk area is a chunk.

Since the data that the information processing apparatus writes for the second time includes the data pattern # 3, the data pattern # 2, and the data pattern # 1, the data pattern # 2 and the data pattern # 1 are deduplicated by the deduplication function. Then, the data pattern # 3 is compressed and added to the chunk in which the compressed data pattern # 1 and the data pattern # 2 are written.

In the variable chunk method in which the chunk size is variable and the deduplication rate is high but the deduplication performance is low, there is a deduplication system that improves the deduplication performance. This deduplication system inputs content from a client terminal, determines a calculation range from a predetermined maximum chunk size and minimum chunk size, and divides the calculation range into at least two small calculation ranges. The deduplication system then sets the position of the window for the rolling hash calculation so that the rolling hash calculation is continuous between the two subcalculation ranges. Then, this deduplication system processes the rolling hash calculation while shifting the window in parallel in at least two subcomputation ranges to cut out chunks from the content. Then, this deduplication system does not store the cut out chunk in the storage device if the chunk having the same contents as the cut out chunk is already stored in the storage device.

In addition, there is a disk array device that can quickly find a suspected failure location. This disk array device includes a plurality of interface modules composed of a plurality of host interfaces and a plurality of disk interfaces, and a switch module for connecting the interface modules. This disk array device is provided in an interface module and includes one or more diagnostic test execution means for performing a write / read test on a memory in another interface module through a switch module. Further, this disk array device includes a diagnostic rule storage means for storing a plurality of diagnostic rules. The diagnostic rule is when a combination pattern of test results of one or more write / read tests in which the pair of the diagnostic test execution means for executing the diagnostic test and the other interface module to be tested are different and the combination pattern is established. Shows the correspondence with the suspected failure location. In addition, this disk array device extracts the diagnostic rule with the smallest number of undetermined diagnostic results required for success / failure determination from the diagnostic rules stored in the diagnostic rule storage means, and determines the success / failure of the extracted diagnostic rule. Select the unexecuted write / read tests required to do this. Then, the disk array device performs a process of instructing the diagnostic test executing means for executing the selected write / read test to execute the test.

International Publication No. 2014/184857 JP-A-2010-277499

In the test of the storage device for reducing the data storage amount shown in FIG. 12, the size of the data written by the storage test device to the storage device and the size of the data stored by the disk 9 are different. Therefore, when testing the storage device, the storage test device calculates the write size for writing data to the entire storage device using the chunk size, the deduplication unit, and the data compression ratio. Here, the deduplication unit is the unit size of the data to be deduplicated by the storage device.

When testing a storage device, the operator investigates the chunk size, deduplication unit, and data compression ratio and defines them in the definition file. However, there is a problem that the test time becomes long due to the investigation of chunk size, deduplication unit and data compression rate.

One aspect of the present invention is to reduce the test time of a storage device.

In one aspect, the storage test apparatus has a first acquisition unit, a second acquisition unit, a first calculation unit, a second calculation unit, and a test execution unit. The first acquisition unit acquires the chunk size, which is a storage allocation unit by the capacity virtualization function of the storage device. The second acquisition unit acquires a deduplication unit, which is a unit for removing data duplication in the storage device. The first calculation unit calculates the data compression rate in the storage device based on the chunk size acquired by the first acquisition unit and the deduplication unit acquired by the second acquisition unit. The second calculation unit calculates the amount of data to be written to the storage device when testing the storage device based on the data compression rate calculated by the first calculation unit. The test execution unit writes the data of the amount of data calculated by the second calculation unit to the storage device, reads it out, and compares the data.

In one aspect, the present invention can reduce the test time of a storage device.

FIG. 1 is a diagram showing a configuration of a test system according to an embodiment. FIG. 2 is a diagram showing a functional configuration of the storage test apparatus. FIG. 3 is a flowchart showing a processing flow by the storage test apparatus. FIG. 4 is a flowchart showing a flow of processing for acquiring chunk size. FIG. 5 is a diagram showing an example of a process for acquiring a chunk size. FIG. 6 is a flowchart showing a flow of processing for acquiring the deduplication unit. FIG. 7A is a diagram showing an example (before deduplication) of the process of acquiring the deduplication unit. FIG. 7B is a diagram showing an example of a process for acquiring a deduplication unit (at the time of deduplication). FIG. 8 is a flowchart showing a flow of processing for calculating the data compression rate. FIG. 9A is a diagram showing an example (light size calculation) of processing for calculating the data compression rate. FIG. 9B is a diagram showing an example (write data creation) of processing for calculating the data compression rate. FIG. 9C is a diagram showing an example of processing for calculating the data compression rate (steps S41 to S48). FIG. 10 is a flowchart showing a flow of processing for performing a test for the entire capacity. FIG. 11 is a diagram showing a hardware configuration of a host that executes a storage test program. FIG. 12 is a diagram for explaining the reduction of the data storage amount.

Hereinafter, examples of the storage test apparatus and the storage test program disclosed in the present application will be described in detail with reference to the drawings. It should be noted that this embodiment does not limit the disclosed technology.

First, the configuration of the test system according to the embodiment will be described. FIG. 1 is a diagram showing a configuration of a test system according to an embodiment. As shown in FIG. 1, the test system 1 according to the embodiment has a host 2 and a storage device 3.

The host 2 is a computer that executes the storage test program 2a that tests the storage device 3, and operates as the storage test device. The storage device 3 is a device that stores data on a disk, and is a device under test that is tested by the storage test program 2a.

The host 2 and the storage device 3 are connected by a host interface and a management LAN (Local Area Network). The host 2 performs I / O access (write and read) to the storage device 3 via the host interface. Further, the host 2 acquires management information such as the total disk capacity and the disk usage amount from the storage device 3 via the management LAN.

The storage device 3 has a disk 31 that stores a logical volume written by the host 2. However, the storage device 3 has a deduplication function, a data compression function, and a capacity virtualization function, and stores data in a disk area having a size smaller than the size of the logical volume written by the host 2. In FIG. 1, one of the data patterns represented by vertical shading is deduplicated, the data pattern represented by vertical shading and the data pattern represented by diagonal shading are compressed and stored in one chunk. Will be done.

The storage device 3 has a management I / F 32. The management I / F 32 acquires management information such as the total disk capacity and the disk usage amount based on the request from the host 2 and responds to the host 2.

Next, the functional configuration of the storage test apparatus will be described. FIG. 2 is a diagram showing a functional configuration of the storage test apparatus. As shown in FIG. 2, the storage test device 20 has a control unit 21 and a test processing unit 22.

The control unit 21 controls the operation of the storage test device 20 based on the operation of the operator. The control unit 21 includes an operator operation processing unit 21a, a test configuration construction processing unit 21b, a message creation processing unit 21c, and a test scheduler 21d.

The operator operation processing unit 21a processes the operation of the storage test device 20 from the terminal device by the operator. Further, the operator operation processing unit 21a outputs a message from each function unit to the terminal device and passes the input value by the operator to the function unit that requests the input value.

The test configuration construction processing unit 21b constructs and deletes the logical volume used in the test for the storage device 3. Further, the test configuration construction processing unit 21b saves the settings of the storage device 3, sets the storage device 3 so as to meet the execution conditions of the test system 1, and returns the saved settings to the storage device 3 after the test.

The message creation processing unit 21c collects information such as messages and test results from each functional unit, and creates a message to be output to the terminal device and the external storage medium.

The test scheduler 21d for the storage device 3 creates a test volume, acquires a chunk size, acquires a deduplication unit, calculates a data compression rate, calculates an amount of data, and performs I / O (write / read / compare). ) Schedule the execution of the process. The write / read / compare is to test the storage device 3 by writing the data to the storage device 3, reading the data, and comparing the data with the written data.

The test processing unit 22 tests the storage device 3 based on the scheduling by the test scheduler 21d. The test processing unit 22 includes a test volume creation unit 22a, a chunk size acquisition processing unit 22b, a deduplication unit acquisition processing unit 22c, a data compression rate calculation processing unit 22d, a data amount calculation processing unit 22e, and a test execution unit. It has 22f and a definition file 22g.

The test volume creation unit 22a creates a volume for the I / O access test by using the entire disk 31 of the storage device 3.

The chunk size acquisition processing unit 22b acquires the chunk size, that is, the allocation unit of the area of the disk 31 by the capacity virtualization function. The chunk size acquisition processing unit 22b receives from the storage device 3 the amount of disk usage after writing 1 byte of data to the storage device 3 in a state where all the data of the storage device 3 is deleted, and sets the chunk size.

The deduplication unit acquisition processing unit 22c acquires the deduplication unit, that is, the unit size of the data to be deduplicated by the storage device 3. The deduplication unit acquisition processing unit 22c gradually increases the size of the data in the process of writing the data from the beginning and the center of the disk 31 in a state where all the data of the storage device 3 is deleted and acquiring the usage amount of the disk 31. Repeat while. Then, the deduplication unit acquisition processing unit 22c sets the size of the data when the usage amount of the disk 31 changes from twice to one times the chunk size as the deduplication unit.

The data compression rate calculation processing unit 22d calculates the data compression rate using the chunk size and the deduplication unit. Specifically, the data compression rate calculation processing unit 22d sets the deduplication unit until the write size is reached by multiplying the chunk size and the least common multiple of the deduplication unit by 1000 in a state where all the data in the storage device 3 is deleted. Write the data so that it is not deduplicated. Then, the data compression rate calculation processing unit 22d acquires the usage amount of the storage device 3, and sets the value obtained by subtracting the ratio of the acquired usage amount to the write size from 1 as the data compression rate. The details of the processing of the data compression rate calculation processing unit 22d will be described later.

The data amount calculation processing unit 22e uses the data compression rate to add the amount of data compression and decrease to the capacity of the disk 31 to perform a test (whole write / read / compare) for the entire capacity of the disk 31. Calculate the amount.

The test execution unit 22f tests the storage device 3 by performing full-scale write / read / compare on the volume on the test system 1 using the data amount calculated by the data amount calculation processing unit 22e.

The definition file 22g stores the chunk size acquired by the chunk size acquisition processing unit 22b, the deduplication unit acquired by the deduplication unit acquisition processing unit 22c, the data compression rate calculated by the data compression rate calculation processing unit 22d, and the like.

Next, the processing flow by the storage test apparatus 20 will be described. FIG. 3 is a flowchart showing a processing flow by the storage test apparatus 20. As shown in FIG. 3, the storage test device 20 saves the current settings of the storage device 3 (step S1).

Then, the storage test apparatus 20 acquires the chunk size (step S2) and acquires the deduplication unit (step S3). Then, the storage test apparatus 20 calculates the data compression rate using the acquired chunk size and the deduplication unit (step S4).

Then, the storage test apparatus 20 calculates the amount of data to be tested (whole surface write / read / compare) for the entire capacity of the disk 31 using the data compression rate, and updates the definition file 22 g (step S5). Then, the storage test device 20 sets the settings of the storage device 3 so as to meet the execution conditions of the test system 1 (step S6).

Then, the storage test apparatus 20 performs write / read / compare for the entire capacity of the disk 31 (step S7). Then, the storage test apparatus 20 writes the area exceeding the total capacity of the disk 31 and confirms that the area exceeding the total capacity of the disk 31 is written by the entire capacity of the disk 31 (step S8).

Then, the storage test device 20 applies the saved setting to the setting of the storage device 3 (step S9). That is, the storage test device 20 returns the setting of the storage device 3 to the shipping setting.

In this way, the storage test apparatus 20 can obtain the chunk size and the deduplication unit and calculate the data compression rate, thereby eliminating the need for the operator to investigate the chunk size, the deduplication unit, and the data compression rate. ..

Next, the flow of the process of acquiring the chunk size will be described. FIG. 4 is a flowchart showing a flow of processing for acquiring chunk size. As shown in FIG. 4, the chunk size acquisition processing unit 22b deletes the data on the disk 31 to be tested (step S21). The chunk size acquisition processing unit 22b deletes the data by, for example, unallocating the data by the UNMAP command of the SCSI (Small Computer System Interface) command.

Then, the chunk size acquisition processing unit 22b acquires the usage amount of the disk 31 and confirms that the usage amount of the disk 31 is 0 (step S22). The chunk size acquisition processing unit 22b acquires the usage amount of the disk 31 by acquiring information from the management LAN.

Then, the chunk size acquisition processing unit 22b writes 1 byte from the beginning of the disk 31 (step S23). By writing only one byte, deduplication and data compression can be avoided. Then, the chunk size acquisition processing unit 22b acquires the usage amount of the disk 31 (step S24). That is, the chunk size acquisition processing unit 22b confirms the usage amount of the disk 31 and acquires it as a chunk size. The chunk size acquisition processing unit 22b acquires the usage amount of the disk 31 by acquiring information from the management LAN.

Then, the chunk size acquisition processing unit 22b deletes the data written in step S23 (step S25). The deletion method is the same as in step S21. Then, the chunk size acquisition processing unit 22b acquires the usage amount of the disk 31 (step S26), and confirms that the usage amount of the disk 31 is 0. The chunk size acquisition processing unit 22b acquires the usage amount of the disk 31 by acquiring information from the management LAN.

FIG. 5 is a diagram showing an example of a process for acquiring a chunk size. As shown in FIG. 5, the chunk size acquisition processing unit 22b deletes the data on the disk 31 in step S21, and acquires the total disk capacity = 10 GB (gigabytes) and the disk usage = 0 MB (megabytes) in step S22. ..

Then, the chunk size acquisition processing unit 22b writes 1 byte in step S23. Since the data is 1 byte, deduplication and data compression are not performed. In addition, the data virtualization function allocates a chunk size = 21 MB of disk 31. Then, the chunk size acquisition processing unit 22b acquires the total disk capacity = 10 GB and the disk usage = 21 MB in step S24, and specifies the chunk size = 21 MB.

In this way, the chunk size acquisition processing unit 22b can specify the chunk size by writing 1 byte in a state where all the data on the disk 31 is deleted and acquiring the usage amount of the disk 31. The number of bytes to be written may be larger than 1 as long as it is certain that one chunk is allocated.

Next, the flow of the process of acquiring the deduplication unit will be described. FIG. 6 is a flowchart showing a flow of processing for acquiring the deduplication unit. It is assumed that the data on the disk 31 has been deleted. As shown in FIG. 6, the deduplication unit acquisition processing unit 22c writes 1 byte from the beginning of the disk 31 (step S31). Then, the deduplication unit acquisition processing unit 22c writes 1 byte of the same data as in step S31 from the center of the disk 31 (step S32).

And. The deduplication unit acquisition processing unit 22c acquires the usage amount of the disk 31 by acquiring information from the management LAN (step S33). Then, the deduplication unit acquisition processing unit 22c determines whether or not the usage amount of the disk 31 is twice the chunk size (step S34).

When the usage amount of the disk 31 is twice the chunk size, the deduplication is not performed. Therefore, the deduplication unit acquisition processing unit 22c increases the size of the data and writes from the beginning of the disk 31 (the deduplication unit acquisition processing unit 22c increases the size of the data). Step S35). The deduplication unit acquisition processing unit 22c gradually increases the size of the data by repeating steps S34 to S37. Further, the deduplication unit acquisition processing unit 22c sets the data as random data so as not to be compressed.

Then, the deduplication unit acquisition processing unit 22c writes the same data as in step S35 by the same size from the center of the disk 31 (step S36). Then, the deduplication unit acquisition processing unit 22c acquires the usage amount of the disk 31 (step S37). The deduplication unit acquisition processing unit 22c acquires the usage amount of the disk 31 by acquiring information from the management LAN. Then, the deduplication unit acquisition processing unit 22c returns to step S34.

On the other hand, when the usage amount of the disk 31 is not twice the chunk size, deduplication has been performed, so that the deduplication unit acquisition processing unit 22c acquires the size of the data written on the disk 31 as the deduplication unit. (Step S38).

FIG. 7A is a diagram showing an example of a process for acquiring a deduplication unit (before deduplication removal), and FIG. 7B is a diagram showing an example of a process for acquiring a deduplication unit (at the time of deduplication removal). As shown in FIG. 7A, the deduplication unit acquisition processing unit 22c writes 1 byte from the beginning of the disk 31 in step S31, and writes 1 byte of the same data from the center of the disk 31 in step S32. Two chunks are assigned to the disk 31.

Then, the deduplication unit acquisition processing unit 22c acquires the total disk capacity = 10 GB and the disk usage = 21 MB × 2 = 42 MB in step S33. Since deduplication has not been performed, disk usage = chunk size x 2.

Then, the deduplication unit acquisition processing unit 22c gradually increases the size of the data to be written, and as shown in FIG. 7B, writes 4KB (kilobytes) from the beginning of the disk 31 in step S35. Then, the deduplication unit acquisition processing unit 22c writes 4KB of the same data from the center of the disk 31 in step S36. Assuming that the deduplication unit is 4KB, deduplication is performed at this point, and one chunk remains on the disk 31.

Then, the deduplication unit acquisition processing unit 22c acquires the total disk capacity = 10 GB and the disk usage = 21 MB in step S37. Since deduplication has been performed, disk usage = chunk size.

In this way, the deduplication unit acquisition processing unit 22c can acquire the deduplication unit by writing the data to the head and center of the disk 31 while gradually increasing the size of the data until the deduplication is performed. .. The place where the two data are written to the disk 31 does not have to be the head and the center as long as it is certain that the two chunks are assigned. The number of writing locations may be three or more.

Next, the flow of processing for calculating the data compression rate will be described. FIG. 8 is a flowchart showing a flow of processing for calculating the data compression rate. It is assumed that the data on the disk 31 has been deleted. Also, in FIG. 8, the write / read / compare time is also calculated.

As shown in FIG. 8, the data compression rate calculation processing unit 22d calculates the write size (write size) required to be written on the disk 31 in order to calculate the data compression rate (step S41). This light size becomes the size to be lighted in steps S42 to S44.

Specifically, the data compression rate calculation processing unit 22d calculates the write size that satisfies the following conditions.
・ Least common multiple of deduplication unit and chunk size ・ 1000 times or more of chunk size

By setting the write size to a multiple of the deduplication unit, the data compression rate calculation processing unit 22d can control the data to be written to the disk 31 for calculating the data compression rate so as not to be deduplicated. Further, by setting the write size to a multiple of the chunk size, the data compression rate calculation processing unit 22d can write the entire area allocated from the disk 31.

The reason why the write size is 1000 times or more the chunk size is that the usage amount of the disk 31 is in chunk size units, but the compressed size is not always in chunk size units, so the compressed size and chunk size. This is to reduce the error from the unit size. By setting the write size to 1000 times or more the chunk size, the error of the data compression rate can be reduced to 1/100 or less.

Then, the data compression rate calculation processing unit 22d writes the size of the deduplication unit as the first light (step S42). Specifically, the data compression rate calculation processing unit 22d prepares the data having the following contents and writes the data for the size of the deduplication unit.
-The first 4 bytes of the data are the time (for example, the cumulative number of seconds since 0:00 on January 1, 1970). This part is used as an area for variable data because deduplication is not possible unless all the data for the size of the deduplication unit match.
-Random data is used after the first 5 bytes of data. Random data is used as data that is difficult to compress.

Then, the data compression rate calculation processing unit 22d creates a data pattern in which the data compression rate is almost the same as that of the first write without deduplication as the second and subsequent writes, and the unit size of the data is deduplicated. Light for minutes (step S43). Specifically, the data compression rate calculation processing unit 22d prepares the data having the following contents, and continuously writes the data at the previously written location.
-The first 4 bytes of the data shall be the data obtained by adding +1 to the first 4 bytes when the previous writing was performed. By changing this part, the data will not be deduplicated.
-The data after the first 5 bytes of the data is the same as the data written the first time. This will compress all the data at about the same compression ratio.

Then, the data compression rate calculation processing unit 22d determines whether or not the total size of the written data has reached the write size calculated in step S41 (step S44), and if not, proceeds to step S43. Return. The data compression rate calculation processing unit 22d repeats the writing in step S43 until the total of the written data reaches the write size.

Then, when the total of the written data reaches the write size, the data compression rate calculation processing unit 22d performs read / compare for the areas written in steps S42 to S44 (step S45). Then, the data compression rate calculation processing unit 22d acquires the writing time in steps S42 to S44 and the read / compared time in step S45 (step S46).

The data compression rate calculation processing unit 22d performs the processing of steps S45 to S46 in order to measure the processing time of write / read / compare. The data compression rate calculation processing unit 22d uses the measured time and the ratio of the write size to the size required to perform the test for the entire capacity of the disk 31 to set the estimated time for performing the test for the entire capacity. Can be calculated.

Then, the data compression rate calculation processing unit 22d confirms the usage amount of the disk 31 and acquires the data amount after data compression (step S47). The data compression rate calculation processing unit 22d specifies the amount of data after data compression by checking the amount of data used by the disk 31. Then, the data compression rate calculation processing unit 22d divides the amount of data acquired in step S47 by the write size calculated in step S41 to calculate the data compression rate (step S48). Data compression rate = (1-disk usage / write size) x 100 (%).

FIG. 9A is a diagram showing an example of a process for calculating the data compression rate (light size calculation), and FIG. 9B is a diagram showing an example of a process for calculating the data compression rate (write data creation), FIG. 9C. Is a diagram showing an example of processing for calculating the data compression rate (steps S41 to S48).

As shown in FIG. 9A, assuming that the chunk size = 21MB and the deduplication unit = 4KB, 4KB × 5376 pieces = 21MB, so that the least common multiple of the chunk size and the deduplication unit is 21MB. Therefore, in step S41, the write size = 21MB × 1000 pieces = 21000MB is calculated.

Further, as shown in FIG. 9B, the first 4 bytes (time) of the data lighted for the first time in step S42 is 0x567774100, assuming that about 46 years have passed since 0:00 on January 1, 1970. The fifth and subsequent bytes (random data) are 0x30289153 ... 79935316. Further, the first 4 bytes (time) of the data to be written second in step S43 is 0x56774101, and the fifth and subsequent bytes (random data) are 0x30289153 ... 799531516.

Further, as shown in FIG. 9C, the data compression rate calculation processing unit 22d calculates the write size in step S41, and writes 4KB of data for 21000 MB in steps S42 to S44. Then, the data compression rate calculation processing unit 22d reads / compares 21000 MB in step S45, and measures the write / read / compare processing time in step S46. Then, the data compression rate calculation processing unit 22d acquires the usage amount of the disk 31 = 19950 MB in step S47, and calculates the data compression rate = (1-1995 50 MB / 21000 MB) × 100% = 5% in step S48.

In this way, the data compression rate calculation processing unit 22d can calculate the data compression rate by calculating the write size and acquiring the usage amount of the disk 31 after writing the data of the deduplication unit for the write size. ..

Next, the flow of the process of performing the test for the total capacity will be described. FIG. 10 is a flowchart showing a flow of processing for performing a test for the entire capacity. As shown in FIG. 10, the data amount calculation processing unit 22e calculates the entire surface write size required for the entire surface write of the disk 31 (step S51). Full write size = total capacity x (1 + data compression rate). For example, assuming that the total capacity = 1000 GB and the compression rate = 5%, the entire surface write size = 1000 GB × 1.05 = 1050 GB.

Then, the data amount calculation processing unit 22e calculates the time required for the entire write / read / compare of the disk 31 from the entire write size calculated in step S51 (step S52). The data amount calculation processing unit 22e is required for full write / read / compare based on the ratio of the write size to the full write size calculated when calculating the data compression rate and the time measured when calculating the data compression rate. Calculate the time. For example, if it takes 1 minute for write and 2 minutes for read / compare when calculating the data compression ratio, the time required for full write / read / compare of 1050 GB is 1050 x 1024 MB / 21000 MB x 3 minutes = 153. 6 minutes. 153.6 minutes is a guideline for testing the entire volume.

Then, the test execution unit 22f writes the data of the entire write size calculated in step S51 from the beginning of the disk 31 (step S53). That is, the test execution unit 22f writes the entire surface of the disc. Then, the test execution unit 22f reads the data written in step S53 (step S54). That is, the test execution unit 22f reads the entire surface of the disk.

Then, the test execution unit 22f compares the data written in step S53 with the data read in step S54 (step S55). Then, the test execution unit 22f determines whether or not the written data and the read data all match (step S56), and if they all match, the test ends normally (step S57). On the other hand, if there is data that does not match, the test execution unit 22f terminates the test abnormally (step S58).

In this way, the data amount calculation processing unit 22e calculates the size required for the entire write of the disk 31, so that the test execution unit 22f can test the entire capacity of the disk 31.

Next, the hardware configuration of the host 2 that executes the storage test program 2a will be described. FIG. 11 is a diagram showing a hardware configuration of the host 2 that executes the storage test program 2a. As shown in FIG. 11, the host 2 has a main memory 51, a CPU (Central Processing Unit) 52, a LAN interface 53, and an HDD (Hard Disk Drive) 54. Further, the computer 50 has a super IO (Input Output) 55, a DVI (Digital Visual Interface) 56, and an ODD (Optical Disk Drive) 57.

The main memory 51 is a memory for storing a program, a result during execution of the program, and the like. The CPU 52 is a central processing unit that reads a program from the main memory 51 and executes it. The CPU 52 includes a chipset having a memory controller.

The LAN interface 53 is an interface for connecting the computer 50 to another computer via a LAN. The HDD 54 is a disk device for storing programs and data, and the super IO 55 is an interface for connecting an input device such as a mouse or a keyboard. The DVI 56 is an interface for connecting a liquid crystal display device, and the ODD 57 is a device for reading and writing a DVD.

The LAN interface 53 is connected to the CPU 52 by PCI Express (PCIe), and the HDD 54 and ODD 57 are connected to the CPU 52 by SATA (Serial Advanced Technology Attachment). The super IO 55 is connected to the CPU 52 by LPC (Low Pin Count).

Then, the storage test program 2a executed on the host 2 is stored in a CD-R, which is an example of a recording medium readable by the host 2, read from the CD-R by the ODD 57, and installed on the host 2. Alternatively, the storage test program 2a is stored in a database or the like of another computer system connected via the LAN interface 53, read from these databases, and installed on the host 2. Then, the installed storage test program 2a is stored in the HDD 54, read into the main memory 51, and executed by the CPU 52.

As described above, in the embodiment, the chunk size acquisition processing unit 22b acquires the chunk size, and the deduplication unit acquisition processing unit 22c acquires the deduplication unit. Then, the data compression rate calculation processing unit 22d calculates the data compression rate based on the chunk size and the deduplication unit, and the data amount calculation processing unit 22e calculates the entire write size based on the data compression rate. Then, the test execution unit 22f tests the storage device 3 by performing full write / read / compare based on the full write size.

Therefore, the operator does not need to investigate the chunk size, the deduplication unit, and the data compression rate when testing the storage device 3, and the storage test device 20 can shorten the test time.

Further, in the embodiment, the chunk size acquisition processing unit 22b acquires the usage amount of the disk 31 after writing 1 byte of data in a state where all the data of the disk 31 is deleted, so that the chunk size is accurate. You can get the size.

Further, in the embodiment, the deduplication unit acquisition processing unit 22c performs a process of writing data to the beginning and center of the disk 31 in a state where all the data of the disk 31 is deleted to acquire the usage amount of the disk 31. Repeat while gradually increasing. Then, the deduplication unit acquisition processing unit 22c acquires the size of the data when the usage amount of the disk 31 changes from twice to one times the chunk size as the deduplication unit. Therefore, the deduplication unit acquisition processing unit 22c can acquire an accurate deduplication unit.

Further, in the embodiment, the data compression rate calculation processing unit 22d deduplications the data of the deduplication unit so that the data of the deduplication unit is not deduplicated until the write size obtained by multiplying the chunk size and the least common multiple of the deduplication unit by 1000 is reached. Write with all the data in. Then, the data compression rate calculation processing unit 22d acquires the usage amount of the disk 31, and calculates a value obtained by subtracting the ratio of the acquired usage amount to the write size from 1 as the data compression rate. Therefore, the data compression rate calculation processing unit 22d can accurately calculate the data compression rate.

Further, in the embodiment, the case where the storage device 3 stores the data on the disk 31 has been described, but the present invention is not limited to this, and the data is stored in another non-volatile storage device such as a NAND flash memory. The same can be applied to the case of.

1 Test system 2 Host 2a Storage test program 3 Storage device 9, 31 Disk 20 Storage test device 21 Control unit 21a Operator operation processing unit 21b Test configuration construction processing unit 21c Message creation processing unit 21d Test scheduler 22 Test processing unit 22a Test volume creation Part 22b Chunk size acquisition processing unit 22c Deduplication unit acquisition processing unit 22d Data compression rate calculation processing unit 22e Data amount calculation processing unit 22f Test execution unit 22g Definition file 32 Management I / F
51 Main memory 52 CPU
53 LAN interface 54 HDD
55 Super IO
56 DVI
57 ODD

Claims (5)

The first acquisition unit that acquires the chunk size, which is the storage allocation unit by the capacity virtualization function of the storage device,
A second acquisition unit that acquires a deduplication unit, which is a unit for removing data duplication in the storage device, and
A first calculation unit that calculates the data compression rate in the storage device based on the chunk size acquired by the first acquisition unit and the deduplication unit acquired by the second acquisition unit.
A second calculation unit that calculates the amount of data to be written to the storage device when testing the storage device based on the data compression rate calculated by the first calculation unit.
A storage test apparatus including a test execution unit for writing data of the amount of data calculated by the second calculation unit to the storage apparatus and then reading and comparing the data. The first acquisition unit uses the usage amount of the storage device as the chunk size after writing data of a predetermined size that is certain to be smaller than the chunk size in a state where all the data of the storage device is deleted. The storage test apparatus according to claim 1, wherein the storage test apparatus is to be acquired. The second acquisition unit acquires the usage amount of the storage device by writing data that is certain to be smaller than the deduplication unit at a plurality of locations distant from the chunk size in a state where all the data of the storage device is deleted. This process is repeated while gradually increasing the size of the data until the usage amount changes from 2 times to 1 time the chunk size, and when the usage amount changes from 2 times to 1 time the chunk size. The storage test apparatus according to claim 2, wherein the size of the data is acquired as the deduplication unit. The first calculation unit performs all of the storage device so that the data of the deduplication unit is not deduplicated until the write size is reached by multiplying the chunk size and the least common multiple of the deduplication unit by a predetermined magnification. The claim is characterized in that the data is written in a deleted state to acquire the usage amount of the storage device, and the value obtained by subtracting the ratio of the acquired usage amount to the writing size from 1 is calculated as the data compression rate. The storage test apparatus according to 3. On the computer
Acquires the chunk size, which is the storage allocation unit, by the capacity virtualization function of the storage device.
A deduplication unit, which is a unit for removing data duplication in the storage device, is acquired.
The data compression ratio in the storage device is calculated based on the chunk size and the deduplication unit.
The amount of data to be written to the storage device when testing the storage device is calculated based on the data compression ratio.
A storage test program characterized in that a process of writing data of the amount of data to the storage device, reading the data, and comparing the data is executed.

Images