JP2002014776A - Disk control system and data rearranging method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make data rearrangement processing for increasing read performance efficient as to a log structured writing type disk control system. SOLUTION: It is decided whether or each logical stripe should be an object of repacking (rearrangement) by referring to an address conversion table. In this decision processing, α showing the rate of the number of effective blocks and β showing how much physical block numbers having adjacent logical block numbers succeed are computed. The repacking processing is carried out for corresponding stripes on condition that 'α is larger than a certain value A and β is less than a certain value B'. At this time, the repacking processing can be carried out preferentially for a logical stripe where many effective blocks are physically dispersed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a disk control system and a data relocation method, and more particularly to a disk control system using a log structured writing method and a data relocation method applied to the system.

[0002]

2. Description of the Related Art In recent years, RAID (Redundant Array) has been developed from the viewpoint of speeding up disks and improving reliability.
2. Description of the Related Art A computer system equipped with a technology of Independent Disks is mainly used. RAID
There are 0, 1, 10, and 5 types of technologies.

Recently, as a technique for greatly improving such RAID write performance, a log-structured file system (LSFS: Log-Structure) has been proposed.
edFile System) technology is being considered.
The log-structured file system is a technique for improving performance utilizing the characteristics of a disk device that "sequential access is much faster than random access". The performance degradation at the time of random access is caused by the seek and rotation wait of the disk, and is a part that should be eliminated if possible. Therefore, in the log-structured file system technology, a method of using a buffer for storing write data, converting the write data of a plurality of small blocks into one large chunk of data, and performing "collective writing". Used. Thus, random writing can be changed to sequential writing, and writing performance can be improved.

[0004]

However, when the log structured file system technique is used, data is arranged and written in the order of write requests. Therefore, when a random write occurs, data whose logical addresses are separated from each other are continuously written in one physical stripe area. This results in data blocks with consecutive logical addresses being written to separate physically separated physical stripe areas on the file system. Sequential read for data stored in such a physically dispersed manner is physically random read and is inefficient.

In order to correct such a state, it is necessary to periodically rearrange the data so that the data is physically continuously arranged. However, the process of rearranging the data itself involves a lot of I / O processing and imposes a heavy load on the system.

The present invention has been made in view of the above circumstances, and has been made to enable efficient data relocation processing.
An object of the present invention is to provide a disk control system and a data rearrangement method capable of greatly improving read performance with a small load.

[0007]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention relates to a disk control system of a log-structured writing system, comprising: a disk array comprising a plurality of disk devices; A data buffer for storing a plurality of data blocks, and write data for one physical stripe of the disk array are generated from the plurality of data blocks stored in the data buffer, and are collectively written to a predetermined physical stripe of the disk array. Data write means for writing data, a plurality of logical block numbers constituting a logical stripe of the disk array, and a physical block number indicating a physical position on the disk array where the data block specified by the logical block number exists. Address translation that stores correspondences for each logical stripe And the address conversion table, and calculates the ratio of the number of valid logical blocks in the logical stripe and / or the ratio of consecutive physical block numbers, and the result satisfies a predetermined condition. In the case, a relocation target determination unit that determines a logical stripe to be relocated, and a data block of the logical stripe whose relocation is determined by the relocation target determination unit is read from the disk array to the data buffer, Data relocation means for writing data to a free physical stripe area of the disk array by data writing means, and address conversion table rewriting means for rewriting the physical block number of the address conversion table to the physical block number of the free physical stripe. It is characterized by the following.

In this disk control system, the degree of physical distribution of logical blocks is checked for each logical stripe by referring to the address conversion table. That is, for each logical stripe, a ratio of the number of valid logical blocks in a plurality of logical blocks constituting the logical stripe and / or a ratio of consecutive physical block numbers of adjacent logical blocks are calculated, and rearrangement is performed based on the calculated ratio. The target logical stripe is determined. Since a logical stripe having a small percentage of the number of valid logical blocks cannot be expected to have a sufficient effect even if rearranged, such a logical stripe is preferably excluded from the rearrangement targets. In addition, some valid logical blocks are included, and
A logical stripe with low continuity of physical block numbers is inefficient at the time of performing sequential read, but the efficiency of sequential read can be improved by making data blocks in such a logical stripe a relocation target. .

The present invention also relates to a disk control system of a log-structured writing system, comprising: a disk array comprising a plurality of disk devices; a data buffer for storing a plurality of data blocks requested to be written; Data write means for generating write data for one physical stripe of the disk array from a plurality of data blocks accumulated in a data buffer and writing the write data to a predetermined physical stripe of the disk array at a time; An address for storing, for each logical stripe, a correspondence relationship between a plurality of logical block numbers constituting a stripe and a physical block number indicating a physical position on the disk array where the data block specified by the logical block number exists. Translation table and the address translation table A counter provided for each logical stripe and counting the number of data reads for the logical block included in the logical stripe; and a logical stripe in which the number of data reads is equal to or more than a predetermined value with reference to the counter. And a data block of the logical stripe for which relocation is determined by the relocation target determining means from the disk array to the data buffer, Data relocation means for writing data to a free physical stripe area of the disk array by means, and an address conversion table rewriting means for rewriting the physical block number of the address conversion table to a physical block number of the free physical stripe. Features .

As described above, by performing data rearrangement on a logical stripe having a large number of data read operations, data blocks in the logical stripe having a high read frequency can be physically contiguous. It is possible to efficiently improve the read performance.

The present invention also relates to a disk control system of a log structured writing system, comprising: a disk array composed of a plurality of disk devices; a data buffer for storing a plurality of data blocks requested to be written; Data write means for generating write data for one physical stripe of the disk array from a plurality of data blocks accumulated in a data buffer and writing the write data to a predetermined physical stripe of the disk array at a time; An address for storing, for each logical stripe, a correspondence relationship between a plurality of logical block numbers constituting a stripe and a physical block number indicating a physical position on the disk array where the data block specified by the logical block number exists. Translation table and the address translation table A counter provided for each logical stripe and counting the number of times of data writing to the logical block included in the logical stripe; A relocation target determining unit that determines a logical stripe to be allocated; and a data block of the logical stripe whose relocation is determined by the relocation target determining unit is read from the disk array to the data buffer, and the data writing unit Data relocation means for writing to a free physical stripe area of the disk array, and address translation table rewriting means for rewriting the physical block number of the address translation table to a physical block number of the free physical stripe. You .

When a logical stripe is written to a logical block constituting the logical stripe, the logical stripe may be physically discontinuous. Therefore, by performing a process for determining a logical stripe to be relocated to a logical stripe whose data write count is equal to or greater than a predetermined value,
It is possible to efficiently improve the performance of sequential access.

[0013]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a configuration of a computer system to which a disk control system according to an embodiment of the present invention is applied. This computer system is a server (PC
Server 1) and a plurality of CPUs 1
1 can be mounted. Each of these CPUs 11 is connected to the bridge 12 via the processor bus 1 as shown. Bridge 12 is connected to processor bus 1 and PCI.
This is a bridge LSI for connecting the bus 2 bidirectionally, and also has a built-in memory controller for controlling the main memory 13. The main memory 13 is loaded with an operating system (OS), an application program to be executed, a driver, and the like.

As shown, the PCI bus 2 has an RAI
The D controller 16 and the nonvolatile memory 17 are connected. The disk array 18 controlled by the RAID controller 16 is used for recording various user data and the like.

The disk array 18 functions as, for example, a disk array having a RAID 5 configuration under the control of the RAID controller 16. In this case, the disk array 18
Is N + 1 disks (here, N + 1 disks for parity storage added to N disks for data storage)
(Disks DISK0 to DISK4). These N + 1 disk devices are grouped and used as a single logical disk drive.

As shown, data (DATA) and its parity (PAR) are stored in the grouped disk devices.
Physical stripes (parity groups) are allocated, and the parity position is sequentially moved between N + 1 disk devices for each physical stripe. For example, in the physical stripe S0, DISK0-D
Parity (PA) of a data (DATA) group on a stripe unit assigned to the same position of each of ISK3
RITY) is recorded on the corresponding stripe unit of DISK4. In the next physical stripe S1, the parity (PARITY) corresponding to the data of the stripe S1 is recorded on the corresponding stripe unit of DISK3. By distributing the parity in N + 1 disk units in units of physical stripes, it is possible to prevent concentration of accesses to the parity disk.

The non-volatile memory 17 is a working storage device used for realizing a log structured file system (LSFS), and includes a battery-backed DRA.
M or a flash EEPROM. The Log Structured File System (LSFS)
It is used to improve the write performance on the disk array 18 having the AID5 configuration. In the present embodiment, the OS file system is not improved, and the log structured file system (LS) is provided by the driver program incorporated in the OS and the nonvolatile memory 17.
FS) has been realized. Here, the principle of write control of the disk array 18 by the log structured file system (LSFS) will be described with reference to FIG.

Log Structured File System (LSFS)
In the writing method using the technology, the data write position is not determined according to the logical address requested by the host (the file system of the OS in the present embodiment), but the write data is sequentially accumulated in the order of the write request from the host. By doing so, a large data block composed of a plurality of write data blocks is formed, and the large data blocks are collectively written to the empty area of the disk array 18 in order from the top. The unit of “collective writing” is a physical stripe. That is, one free physical stripe is generated each time “collective writing” is performed, and data blocks for one physical stripe are sequentially written therein. As a result, random access can be converted into sequential access, and the write performance can be greatly improved.

FIG. 2 shows that the block data size of the write data sent via the OS file system is 2 KB and the data size of the stripe unit is 64 K.
B is an example where the data size of one physical stripe (parity group) is 256 KB (= 64 KB × 4). The 2 KB write data block is acquired by the RAID speed-up driver 100 incorporated in the OS and stored in the nonvolatile memory 17. RAID
The high-speed driver 100 is a driver program for realizing a log structured file system (LSFS).

Basically, when 256 KB data blocks (2 KB × 128 data blocks) are stored in the nonvolatile memory 17, one physical block of the disk array 18 is controlled under the control of the RAID acceleration driver 100. Writing to the stripe is performed at once. In this case, the RAID controller 16 can generate parity only from the 256 KB write data block, so that processing such as reading old data for parity calculation becomes unnecessary, and the well-known RAID5 write penalty can be reduced. it can.

As shown in FIG. 1, the nonvolatile memory 17 has a log memory (batch writing buffer) 17.
1, an address conversion table 172, a logical stripe management table 173, and the like are mounted, and access is controlled by a memory control unit (not shown).

The log memory 171 is a data buffer for storing write data blocks. When one physical stripe of write data blocks is stored in the log memory 171, batch write to the disk array 18 is started. You. That is, the physical stripe is a unit of “collective writing”, and is made up of a continuous area on a partition generated for LSFS in the entire storage area of the disk array 18. Each physical stripe has a size that is an integral multiple of the stripe unit managed by the RAID controller 16.

The address conversion table 172 has the LSFS
Relationship between each of a plurality of logical block numbers constituting a logical address space of a partition used for use and a physical block number indicating a physical position on the disk array 18 where the data block specified by the logical block number exists. Is stored.
Here, the logical block number is the number of the logical block on the disk partition as viewed from the file system of the OS. The block number when the file system of the OS requests access is a logical block number, which is a virtual logical address. This logical block number is associated with a physical block number (a physical block number on a disk partition) by the address conversion table 172. The byte offset value from the head position of the disk partition is obtained from the physical block number × block size (byte).

The address conversion table 172 has a plurality of entries corresponding to the respective logical block numbers. When writing a new data block, the physical block number (physical address) where the data block is actually written is registered in the entry corresponding to the logical block number for which the write request was made. When reading a data block, the physical block number is checked from the entry corresponding to the logical block number requested to be read,
A data block is read from a physical position on the disk partition specified by the physical block number (physical block number × block size).

The logical stripe management table 173 is for managing information on each logical stripe, and the information of each logical stripe is used for processing for data relocation. Here, the logical stripe is a set of logical blocks when the logical address space of the disk partition is divided from the head by the number of data blocks of one physical stripe. For example, if one physical stripe includes 10 physical blocks,
One logical stripe also includes ten logical blocks.

As described above, in the writing method of this embodiment, random writes are collectively written into a physical stripe as one continuous area. As a result, even if the randomly written area is continuous data in the logical address space on the OS file system, the data is included in separate physically separated physical stripes. Sequential reading for data written in different physical stripes in this manner is physically random reading and is inefficient. In order to correct such a state, it is necessary to periodically rearrange the data blocks so that the data is physically continuously arranged. This data rearrangement process, that is, "data rearrangement for the purpose of efficiently performing sequential read" is herein referred to as repacking process.

Hereinafter, the control method of the repacking process in the first embodiment will be specifically described with reference to FIGS. FIG. 3 shows an example of the address conversion table 172, and FIG. 4 is a flowchart of the repacking control process.

The address conversion table 172 in FIG. 3 assumes a case where a physical stripe (also a logical stripe) is composed of ten data blocks. The address conversion table 172 includes a large number of entries each having a logical block number as an index, and a physical block number is registered in each entry. "-" Indicates an invalid block (a block in which data has never been written yet).

In the example of the address conversion table 172, one logical stripe is formed for every ten logical blocks from the top. For example, a logical stripe 0 is composed of ten logical blocks having logical block numbers 0 to 9,
Logical stripe 1 is logical block number 10 to 19
It consists of zero logical blocks. That is, when one physical stripe and one logical stripe are each composed of ten data blocks, the relationship of logical stripe number = logical block number / 10 physical stripe number = physical block number / 10 is established. Here, “/” indicates integer division (rounded down to the decimal point).

Focusing on logical stripe 0, the data block of logical block number 0 is the physical block number 30.
(The first data block of the physical stripe 3). Similarly, the data block of the logical block number 1 is the physical block number 31 (the second data block of the physical stripe 3), and the data block of the logical block number 2 is the physical block. It can be seen that the data block of the logical block number 3 is stored in the physical block number 33 (the fourth data block of the physical stripe 3), while the data block of the logical block number 3 is stored in the physical block number 33 (the third data block of the physical stripe 3). Further, data blocks corresponding to the logical block numbers 6, 7, and 8 are physical block numbers 133, 134, and 2, respectively.
37 respectively. Logical block number 4,
5 and 9 are invalid blocks.

As described above, the address conversion table 172
, It is possible to detect the degree of physical distribution of logical blocks constituting each logical stripe for each logical stripe. Therefore, in the present embodiment, by referring to the address conversion table 172 and checking the physical block numbers of the logical blocks constituting each logical stripe, the ratio of the effective block number (=
α), the ratio (= β) of consecutive physical block numbers of adjacent logical blocks is calculated for each logical stripe. The values of α and β determine whether or not the logical stripe is to be repacked.

(Example of Calculation of α and β) Here, an example of calculation of α and β will be described. In FIG. 3, logical stripe 0
Focusing on, since the number of effective blocks is 7, α of the logical stripe 0 is calculated by α = 7/10 × 100 = 70%. The value of α indicates the capacity usage rate of the target logical stripe. The smaller the value, the more blocks that have not been written, and it can be determined that the blocks should be excluded from the repacking target (there are few blocks to be rearranged even after repacking and the effect cannot be expected).

In the logical stripe 0, the physical block numbers of the logical blocks 0, 1, 2, 3 are continuous,
The physical block numbers of the logical blocks 6 and 7 are continuous. For this reason, four boundaries among nine block boundaries in total are physically continuous. Therefore, β of the logical stripe 0 is calculated by β = 4/9 × 100 ≒ 44%.

Similarly, α = 80% and β = 22% for the logical stripe 1, α = 20% and β = 0% for the logical stripe 2, and α for the logical stripe 3.
= 90% and β = 44%.

(Judgment of repacking) Whether or not to be repacked for each logical stripe is determined by using “condition to be repacked when α is greater than constant value A and β is less than or equal to constant value B” (condition 1). Is determined. Where A = 50%,
B = 50%.

In this case, since the value of α of the logical stripe 2 is small, it does not meet the above condition, and it is determined that repacking is unnecessary. This is because "there are few effective blocks and the effect of rearrangement cannot be expected." The logical stripes 0, 1, and 3 meet the above conditions and are determined to be repacked. Because of "the number of effective blocks is sufficient and the continuity is low", it is subject to repacking.

In practice, α and β of each logical stripe change as the repacking process is performed. Therefore, in the present embodiment, the repacking process is sequentially performed for each logical stripe determined to be a repacking target, and the calculation of α and β for the next logical stripe is performed after the repacking process is completed.

(Procedure of Repack Control Process) The procedure of the repack control process will be described below with reference to the flowchart of FIG. This repacking control process is performed by the RAID acceleration driver 100.

First, the address conversion table 172 is referred to in index order, and the value of the corresponding physical block number is checked (step S11). Then, the logical stripes having smaller values are selected in order (step S12), and the following “repacking target stripe selection processing & repacking processing (* 1)” is executed on the selected logical stripes.

In the repacking target stripe selecting process & repacking process, first, α and β for the selected first logical stripe are calculated (step S13). The value of the number of effective blocks for each logical stripe is the number of read references,
The information is managed for each logical stripe in the logical stripe management table 173 together with information such as the number of times of writing. The logical stripe management table 173 is saved together with the address translation table 172 in a separate area from the data area in the LSFS disk partition when the computer system is shut down, and is stored in a non-volatile area from the disk partition when the computer system is booted. The memory is used by being loaded into the memory 17. The value of α is
It can be calculated based on the number of valid blocks in the logical stripe management table 173. Of course, as described above, by registering the information “−” indicating the invalid block in the address conversion table 172, α can be easily obtained only by searching for an entry in the address conversion table 172.

If "α is equal to or less than the fixed value A" (YES in step S14), the current logical stripe is excluded from the repacking target. On the other hand, if “α is larger than the fixed value A” (NO in step S14),
It is determined whether or not the following is true (step S15).
If “α is larger than the fixed value A and β is equal to or smaller than the fixed value B” (YES in step S15), the current logical stripe is determined as a repacking target, and repacking processing is performed on the logical stripe. (Step S
16). The repacking process is a process of rearranging data blocks having consecutive logical block numbers so that they are also physically continuous. In this repacking process, while data blocks of a logical stripe to be repacked are read from the disk array 18 to the log memory 171, a free physical stripe is prepared on a partition of the disk array 18, and read out to the log memory 171 there. The written data blocks are collectively written in the order of the logical block numbers.

Next, it is determined whether or not the processing has been completed for all the logical stripes (step S1).
7) Until there are no unprocessed logical stripes, the repacking target stripe selection processing and the repacking processing are sequentially performed in index order.

The repacking control process described above may be executed automatically, for example, at regular intervals, or may be executed in response to an explicit operation instruction from a system administrator.

FIGS. 6 and 7 show an example of the result of repacking the logical stripe 3 of FIG. FIG. 6 shows the contents of the address conversion table 172 before and after repacking for the logical stripe 3,
FIG. 7 shows the arrangement of data blocks on the physical stripe before and after repacking for the logical stripe 3.

Before repacking, logical stripe 3
9 effective blocks (logical block numbers 30, 3)
1,... 38) are distributed and stored in three physical stripes (physical stripes 6, 11, 15). Note that reference numerals 30 to 38 in FIG. 7 mean data of the logical block number. If the logical blocks 30 to 38 are successively read in this state, three physical stripes 6, 11, and 15 that are apart from each other are actually accessed, which is inefficient. The repacking is performed on the logical stripe 3. Now, the empty physical stripe 1 on the disk array 18
Assuming that 0 is selected, the data of all the valid blocks of the logical stripe 3 is
The data is read out from the log memories 171 and 15 to the log memory 171, and is collectively written to the physical stripe 10. in this case,
In the physical stripe 10, all nine effective blocks are arranged in the order of logical block numbers (30, 31,... 38). In accordance with this, in the address translation table 172, the physical block numbers of the nine valid blocks (logical block numbers 30, 31,... 38) in the logical stripe 3 are updated (100, 10).
1, ... 108). By the repacking process, when the logical blocks 30 to 38 are continuously read, only the access within the physical stripe 10 is performed, and efficient access is realized. As a result, the sequential read performance can be improved.

Further, a new physical stripe is generated for each logical stripe to be repacked, and the data blocks included in the logical stripe to be repacked are arranged and arranged in the order of the logical block numbers, thereby repeating the repacking process. Thus, the physical data arrangement comes closer to the logical data arrangement. Furthermore, by arranging them so that they are physically continuous even in units of logical stripes, it is possible to achieve continuity of data blocks even at physical stripe boundaries. In the repacking process, it is most preferable to arrange the data blocks in the order of the logical block numbers. However, it is a matter of course that the read performance can be improved only by accommodating all the data blocks belonging to the logical stripe to be repacked in the same physical stripe.

Further, a condition that “if β is equal to or less than a fixed value, the repacking target” may be added, and a logical stripe to be repacked may be selected from β alone.

(Repacking Selection Process Based on Read Reference Count) Next, a second embodiment of the repacking process will be described with reference to FIGS. Here, the number of read references is managed for each logical stripe, and the “repacking target stripe selection process” is executed preferentially for logical stripes in which the number of read references exceeds a certain value. The value of the read reference count of each logical stripe means the read reference count after the last repacking stripe selection process for the logical stripe is performed. For this reason, the value of the read reference count of each logical stripe is cleared to 0 each time the repacking target stripe selection processing of the logical stripe is performed. In the case of a logical stripe that has never been the target of the repack selection process, the number of times is "the number of read references since the partition generation time".

FIG. 8 shows a case where RAID is read when reading a data block.
5 is a flowchart illustrating a processing procedure executed by a speed-up driver. Here, a case is assumed in which a read reference counter is prepared for each logical stripe, and the read reference count of each logical stripe is managed using the read reference counter. Referenced_Strip
e_Table [] is a table of “read reference counter”, which has a logical stripe number as an index and a “read reference counter value” as a value. Initialize with 0 at the time of partition generation.

As shown in FIG. 8, at the time of reading a data block, first, the logical stripe number L to which the logical block number is read from the file system of the OS is obtained (step S21). Assuming that the logical block number to be read is Vn and the total number of blocks in each logical stripe is Nb, L is given by L = Vn / Nb.

Next, Referenced_Stri
“read reference counter” of the logical stripe L to which the target logical block belongs as pe_Table [L] ++
Is incremented (step S22). After that,
Normal read processing is performed (steps S23 and S2).
4). That is, the physical block number Pn corresponding to the logical block number Vn is obtained with reference to the address conversion table 172 (step S23), and the physical position on the disk partition specified by the disk address given by Pn × block size , The corresponding data block is read through the RAID controller 16 (step S24).

FIG. 9 is a flowchart of a process executed by the RAID high-speed driver 100 during the repacking process. In FIG. 4, the contents of the address conversion table 172 are referred to in the order of the index.
“Repacking target stripe selection process” using α and β for a logical stripe having a large number of read references using enhanced_Stripe_Table [].
Is performed.

That is, first, the logical stripe numbers are sorted in descending order by the value of the “read reference counter”, and the repacking target stripe is selected from the logical stripe having the larger value of the read reference counter (step S 31). . In step S31, a new array A
[] Is made. Each entry of array A [] has two members, Logical_Stripe_ID, and R
E_reference_Count, expressed in C language, A []. Logical_Stripe_ID A []. It is represented as Reference_Count. A []. Logical_Str
IPe_ID is a logical stripe number (Reference)
ed_Stripe_Table []), and A []. Reference
_Count indicates the number of read references. Referen
By sorting the values registered in ced_Stripe_Table [], the "read reference counter"
Are set in the two members of the array A, respectively, in ascending order of their values and the number of the logical stripe.

Thereafter, the entries are referred to in the order registered in the array A [], and the read reference counter “A [] .R” is read.
As long as there is a logical stripe for which “reference_Count” is equal to or more than the fixed value C, “repacking target stripe selection processing & repacking processing (processing of * 1 in FIG. 4)” is performed. That is, first, assuming that n = 0, the logical stripe having the largest number of read references (L0 = A [0] .Logical).
_Strip_ID) (steps S32 and S32).
33), and its read reference count (A [0] .Ref)
It is determined whether or not (rence_Count) is equal to or smaller than a fixed value C (step S34). If the number of read references is greater than the fixed value C, the above-described repacking target stripe selection processing & repacking processing based on α and β is executed (step S35). Regardless of whether or not the logical stripe is selected as a repacking target, the read reference counter of the corresponding logical stripe is cleared to 0 at the time of the repacking target stripe selection processing (step S36).

Thereafter, as n + 1, the logical stripe having the second largest number of read references (L1 = A [1] .Logi).
cal_Strip_ID) (step S3
7, S38, S33) and its read reference counter (A
[1]. Reference_Count) is a constant value C
It is determined whether or not it is below (step S34).

The read reference counter "A [n] .Ref"
When a logical stripe whose “rence_Count” is equal to or smaller than the fixed value C appears (YES in step S34), the processing is ended. Therefore, a logical stripe having a large number of read references can be preferentially targeted for repacking. Also, the read reference counter “A [] .Reference”
Since it is sufficient to perform the repacking target stripe selection process only for the logical stripes whose _Count is equal to or greater than the fixed value C, the load required for the process can be reduced.

(Repack Selection Process Based on Number of Writes) Next, a third embodiment of the repack process will be described with reference to FIGS. Here, the number of times of writing of the logical block belonging to each logical stripe is managed, and the “repacking target stripe selection process” is preferentially executed for a logical stripe whose number of times of writing has exceeded a certain value. The value of the number of times of writing of each logical stripe means the number of times of writing after the last repacking stripe selection process for the logical stripe is performed. For this reason, the value of the number of times of writing of each logical stripe is cleared to 0 every time the repacking target stripe selection processing of the logical stripe is performed. In the case of a logical stripe that has never been subjected to the repack selection process, the number of times of writing is “the number of times of writing since the partition was created”.

When a logical stripe is written to a logical block constituting the logical stripe, there is a possibility that the logical stripe becomes physically discontinuous. This is because, in LSFS, additional writing processing is performed in units of physical stripes, and old data blocks on another physical stripe corresponding to data blocks included in a newly written physical stripe are invalidated. Therefore, it is considered that the possibility that the physical position where the logical block exists is dispersed is higher in the logical stripe having the larger number of times of writing. Therefore,
Efficient repacking processing can be realized by preferentially setting a logical stripe having a large number of writing times as a candidate for repacking processing.

FIG. 10 is a flowchart showing a processing procedure executed by the RAID acceleration driver 100 when writing data. Here, it is assumed that a write frequency counter is prepared for each logical stripe and the write frequency counter is used to manage the write frequency of each logical stripe. Modified_Stripe_T
able [] is a table of a “write count counter”, and uses a logical stripe number as an index.
It has a “write counter value” as a value. Initialize with 0 at the time of partition generation.

As shown in FIG. 10, when data blocks for one physical stripe are accumulated in the log memory 171,
The data blocks for the one physical stripe are collectively written to the physical stripe to be written (step S41). Next, the logical stripe number L to which the written data block belongs is obtained (step S42). The logical block number of each data block is V
n, and the total number of blocks in the logical stripe is Nb.
The logical stripe number L is given by L = Vn / Nb.

Thereafter, the Modified_Stripe
As “_Table [L] ++”, the “write number counter” of the logical stripe to which the target logical block belongs is incremented (step S43). Thereafter, as part of the normal write processing, the physical block number Pn actually written is set in the entry of the address conversion table 172 corresponding to the logical block number Vn for each written data block ( Step S44). Steps S41 to S44 described above are executed for all blocks in the physical stripe.

FIG. 11 is a flowchart of a process executed by the RAID high-speed driver 100 during the repacking process. In FIG. 4, the contents of the address conversion table 172 are referred to in the order of the index.
The “repacking target stripe selection process” using α and β described above is performed on a logical stripe having a large number of write operations using the modified_Stripe_Table [].

That is, first, the logical stripe numbers are sorted in descending order by the value of the "write count counter", and the repacking target stripe selection process is performed from the logical stripe having the largest write count counter (step S51). In step S51, a new array A
[] Is made. Each entry in array A [] has two members, Logical_Stripe_ID, and M
modified_Count, expressed in C language, A []. Logical_Stripe_ID A []. Expressed as Modified_Count. A []. Logical_Str
The type_ID is a logical stripe number (Modified)
_Strip_Table []), and A []. Modified_Co
unt indicates the number of times of writing. Modified_St
By sorting the values registered in “ripe_Table []”, the values of the “write count counter” and the number of the logical stripe are set in the two members of the array A in ascending order.

Thereafter, the entries are referred to in the order registered in the array A [], and the write counter “A [].
As long as there is a logical stripe whose “Modified_Count” is equal to or greater than the fixed value D, “repacking target stripe selection processing & repacking processing (processing of * 1 in FIG. 4)” is performed. That is, first, assuming that n = 0, the logical stripe having the largest number of write operations (L0 = A [0] .Logical_
Focusing on (Strip_ID) (Steps S52 and S5)
3), the write counter (A [0] .Modi)
It is determined whether or not (feed_Count) is equal to or smaller than a fixed value D (step S54). If the number of times of writing is larger than the fixed value D, the aforementioned repacking target stripe selection processing and repacking processing based on α and β are executed (step S55). Regardless of whether or not the logical stripe is selected as a repacking target, the write count counter of the corresponding logical stripe is cleared to 0 during the repacking target stripe selection processing (step S55).

Thereafter, as n + 1, a logical stripe having the second largest number of write operations (L1 = A [1] .Logic)
al_Stripe_ID) (step S5).
7, S58, S53), the number of times of writing (A
[1]. It is determined whether or not (Modified_Count) is equal to or smaller than a fixed value D (step S54).

The number-of-writes counter “A [n] .Mod
When a logical stripe whose “ifed_Count” is equal to or smaller than the fixed value D appears (YES in step S54), the process ends. Therefore, it is possible to prioritize a logical stripe having a large number of write operations as a repacking target. Also, the write counter “A [] .Modified_Co
"unpack" may be performed only on logical stripes having a predetermined value D or more.
The load required for the processing can be reduced.

In the second and third embodiments, the repacking target selection process based on α and β in the first embodiment should be performed with priority given the values of the number of read references and the number of writes. Although the logical stripe is determined, the logical stripe to be repacked is determined only by controlling the value of the read reference count and the value of the write count, or both the read reference count value and the write count value are used. Thus, a sufficient effect can be obtained even when the logical stripe to be repacked is determined.

As described above, the present embodiment is most effective when applied to a disk array having a RAID configuration. However, in a disk control system using a log structured writing method, a disk drive composed of one disk drive can be used. The same can be applied to control.

Further, by recording a computer program including the processing procedure of the RAID acceleration driver 100 on a computer-readable recording medium, the present invention can be implemented simply by introducing the computer program to a normal computer system through the recording medium. The same effect as in the embodiment can be obtained.

Further, the present invention is not limited to the above-described embodiment, and can be variously modified in an implementation stage without departing from the gist thereof. Further, the embodiments include inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some components are deleted from all the components shown in the embodiment, the problem described in the column of the problem to be solved by the invention can be solved, and the effects described in the column of the effect of the invention can be solved. If you get
A configuration from which this configuration requirement is deleted can be extracted as an invention.

[0071]

As described above, according to the present invention,
Data can be rearranged efficiently, and read performance can be improved with a small load. In particular, by using the ratio of the number of valid blocks and the ratio of continuous physical block numbers of blocks having adjacent logical block numbers, it is possible to efficiently select logical stripes that need to be rearranged.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a computer system according to an embodiment of the present invention.

FIG. 2 is an exemplary view for explaining the principle of a writing process to a disk device provided in the computer system of the embodiment.

FIG. 3 is an exemplary view showing an example of the contents of an address conversion table provided in the computer system of the embodiment.

FIG. 4 is an exemplary flowchart illustrating a procedure of a repacking control process executed by the computer system of the embodiment.

FIG. 5 is an exemplary view showing the configuration of a logical stripe management table provided in the computer system of the embodiment.

FIG. 6 is an exemplary view showing how the contents of the address conversion table change before and after the repacking process in the computer system of the embodiment.

FIG. 7 is an exemplary view showing how the arrangement of physical data changes before and after the repacking process in the computer system of the embodiment.

FIG. 8 is an exemplary flowchart showing the processing procedure at the time of data reading in the computer system of the embodiment.

FIG. 9 is an exemplary flowchart showing a second procedure of the repacking control process executed by the computer system of the embodiment.

FIG. 10 is an exemplary flowchart showing the processing procedure at the time of data writing in the computer system of the embodiment.

FIG. 11 is an exemplary flowchart illustrating a third procedure of the repacking control process executed by the computer system of the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 ... CPU 12 ... Bridge 13 ... Main memory 16 ... RAID controller 17 ... Non-volatile memory 18 ... Disk array 171 ... Log memory 172 ... Address conversion table 173 ... Logical stripe management table

Claims (10)

[Claims] 1. A disk control system of a log structured write system, comprising: a disk array composed of a plurality of disk devices; a data buffer for storing a plurality of data blocks requested to be written; Data write means for generating write data for one physical stripe of the disk array from the plurality of data blocks thus written, and writing the write data collectively to a predetermined physical stripe of the disk array; and forming a logical stripe of the disk array An address conversion table that stores, for each logical stripe, a correspondence relationship between a plurality of logical block numbers and a physical block number indicating a physical position on the disk array where the data block specified by the logical block number exists; Referring to the address conversion table, The ratio of the number of valid logical blocks in the logical stripe and / or the ratio of consecutive physical block numbers are calculated, and if the result satisfies a predetermined condition, the logical stripe is determined as a logical stripe to be relocated. A relocation target determination unit; and a data block of the logical stripe, the relocation of which is determined by the relocation target determination unit, being read from the disk array to the data buffer, and a free physical stripe area of the disk array being read by the data writing unit. And a data relocation unit for rewriting the physical block number of the address translation table with a physical block number of the empty physical stripe. 2. The relocation target determining unit re-creates a logical stripe in which the ratio of the valid logical blocks is larger than a predetermined value and the ratio of consecutive adjacent physical block numbers is equal to or smaller than a predetermined value. 2. The disk control system according to claim 1, wherein a condition is determined as an arrangement target. 3. A disk control system of a log-structured writing system, comprising: a disk array composed of a plurality of disk devices; a data buffer for storing a plurality of data blocks requested to be written; Data write means for generating write data for one physical stripe of the disk array from the plurality of data blocks thus written, and writing the write data collectively to a predetermined physical stripe of the disk array; and forming a logical stripe of the disk array An address conversion table that stores, for each logical stripe, a correspondence relationship between a plurality of logical block numbers and a physical block number indicating a physical position on the disk array where the data block specified by the logical block number exists; Each logical list in the address translation table A counter for counting the number of data reads for the logical block included in the logical stripe provided for each type, and referring to the counter for re-reading the logical stripe having the number of data reads equal to or more than a predetermined value. A relocation target determining unit that determines a logical stripe to be relocated; a data block of the logical stripe for which relocation is determined by the relocation target determining unit is read from the disk array to the data buffer; Data relocation means for writing to a free physical stripe area of the disk array; and address translation table rewriting means for rewriting the physical block number of the address translation table to the physical block number of the free physical stripe. Diss Control system. 4. A disk control system of a log structured write system, comprising: a disk array composed of a plurality of disk devices; a data buffer for storing a plurality of data blocks requested to be written; Data write means for generating write data for one physical stripe of the disk array from the plurality of data blocks thus written, and writing the write data collectively to a predetermined physical stripe of the disk array; and forming a logical stripe of the disk array An address conversion table that stores, for each logical stripe, a correspondence relationship between a plurality of logical block numbers and a physical block number indicating a physical position on the disk array where the data block specified by the logical block number exists; Each logical list in the address translation table Counter means for counting the number of times of data writing to the logical block included in the logical stripe, provided for each type, and referring to the counter means, the logical stripe whose data writing number is equal to or more than a predetermined value is to be relocated. A relocation target determining unit that determines the logical stripe of the logical stripe; a data block of the logical stripe whose relocation is determined by the relocation target determining unit is read from the disk array to the data buffer; Data relocation means for writing to a free physical stripe area of an array; and address conversion table rewriting means for rewriting the physical block number of the address conversion table to the physical block number of the free physical stripe. This Control system. 5. The reallocation target determining means calculates a ratio of the number of valid logical blocks in the logical stripe and / or a ratio of consecutive physical block numbers in the logical stripe with reference to the address conversion table. 5. The disk control system according to claim 3, further comprising means for determining a logical stripe to be relocated when the result satisfies a predetermined condition. 6. A data relocation method for relocating data blocks stored in a disk array composed of a plurality of disk devices, wherein a plurality of data blocks requested to be written are stored in a data buffer. Generating write data for one physical stripe of the disk array from the plurality of data blocks stored in the data buffer, and writing the data in a predetermined physical stripe of the disk array at once; An address conversion table indicating a correspondence relationship between a plurality of logical block numbers constituting a logical stripe of the array and a physical block number indicating a physical position on the disk array to which a data block of the logical block number is written is provided for each logical stripe. Generating the address, and the address With reference to the conversion table, the ratio of the number of valid logical blocks in the logical stripe and / or the ratio of consecutive physical block numbers are calculated, and if the result satisfies a predetermined condition, the reallocation is performed. Deciding as a target logical stripe; reading data blocks of the logical stripe for which relocation has been decided from the disk array to the data buffer, and writing the data to a free physical stripe area of the disk array by the batch write step Performing a reallocation, and rewriting the physical block number of the address translation table with a physical block number of the empty physical stripe in accordance with the data reallocation. . 7. A condition that a logical stripe whose effective logical block ratio is larger than a predetermined value and whose adjacent physical block number is continuous is equal to or lower than a predetermined value is conditionally determined as a relocation target. 6. The data rearrangement method according to claim 5, wherein: 8. A data relocation method for relocating data blocks stored in a disk array composed of a plurality of disk devices, wherein a plurality of data blocks requested to be written are stored in a data buffer. Generating write data for one physical stripe of the disk array from the plurality of data blocks stored in the data buffer, and writing the data in a predetermined physical stripe of the disk array at once; An address conversion table indicating a correspondence relationship between a plurality of logical block numbers constituting a logical stripe of the array and a physical block number indicating a physical position on the disk array to which a data block of the logical block number is written is provided for each logical stripe. Generating the logical list; For each type, a step of counting the number of data reads for the logical block included in the logical stripe, and a step of determining a logical stripe whose number of data reads is equal to or greater than a predetermined value as a logical stripe to be relocated, Reading a data block of the logical stripe for which relocation is determined from the disk array to the data buffer, writing the data block to a free physical stripe area of the disk array by the batch write step, and relocating the data; Rewriting the physical block number of the table to the physical block number of the empty physical stripe. 9. A data relocation method for relocating data blocks stored in a disk array constituted by a plurality of disk devices, wherein a plurality of data blocks requested to be written are stored in a data buffer. Generating write data for one physical stripe of the disk array from the plurality of data blocks stored in the data buffer, and writing the data in a predetermined physical stripe of the disk array at once; An address conversion table indicating a correspondence relationship between a plurality of logical block numbers constituting a logical stripe of the array and a physical block number indicating a physical position on the disk array to which a data block of the logical block number is written is provided for each logical stripe. Generating the logical list; Counting the number of times data is written to the logical block included in the logical stripe for each type; determining a logical stripe whose data write number is equal to or greater than a predetermined value as a logical stripe to be relocated; Reading the data blocks of the logical stripe determined from the disk array into the data buffer, writing the data blocks to the free physical stripe area of the disk array by the batch write step, and relocating the data, Rewriting the physical block number to the physical block number of the empty physical stripe. 10. The method of determining a logical stripe to be relocated, comprising the steps of:
A ratio of the number of valid logical blocks in the logical stripe and / or a ratio of consecutive physical block numbers are calculated, and if the result satisfies a predetermined condition, the logical stripe is determined as a logical stripe to be relocated. 10. The data relocation method according to claim 8, further comprising steps.