JP3987295B2 - Processing optimization method, computer, and storage device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a processing optimization method, a computer, and a storage device, and more particularly, to a system including a plurality of storage devices that perform normal processing according to an access request from a computer and other background processing. The present invention relates to an applied processing optimization method, a computer, and a storage device.
[0002]
[Prior art]
Many conventional storage devices perform processing (normal processing) in response to an access request from application software running on a host and other processing (background processing).
[0003]
Background processing is, for example, disk repair operation when a failure occurs in an HDD in a storage device, internal data check, data transfer between storage devices, etc., and normal processing according to an access request from application software It means processing other than.
[0004]
Conventionally, when both the normal process and the background process are performed, the background process is performed within a range that does not affect the normal operation or at a fixed priority. This was to make one storage device usable for a plurality of purposes (generally usable).
[0005]
[Problems to be solved by the invention]
In recent years, in a system composed of a plurality of host computers and storage devices, the roles of individual storage devices in the system have been differentiated. Along with this, there are multiple storage devices in the system, roles are divided in the system, temporary buffers for performing backup to secondary storage media such as operation system, standby system, primary backup, tape device etc. It is desired to be able to be used for various purposes.
[0006]
However, the priority of each background process is only controlled at a uniform priority so as not to affect the normal operation as much as possible, and when it becomes necessary to perform a plurality of background processes, these background processes are performed. There is a problem that the ground processing cannot be executed in an optimal order according to the importance at an arbitrary time.
[0007]
In addition, if the usage or status of these storage devices changes at an arbitrary point in time, the priority between normal operation and multiple background processes cannot be changed dynamically. As a result, it is impossible to optimize the processing of the entire system.
[0008]
The present invention has been made in view of the above circumstances, and in a system including a plurality of storage devices that perform normal processing in response to an access request from a computer and a plurality of other background processing, processing of the entire system It is an object to provide a process optimization method, a computer, and a storage device that can optimize the process.
[0009]
[Means for Solving the Problems]
The processing optimization method according to the present invention includes a computer that issues an access request to an arbitrary storage device, and a plurality of storage devices that perform normal processing and other background processing in response to the access request from the computer. A process optimization method applied to a system comprising: a computer for determining a use of each logical unit in each storage device based on an information table, and for each logical unit based on a result of the determination; The priority table indicating the processing priority between the normal process and the plurality of background processes is created for each storage device and issued to each storage device. A priority table is acquired and stored, and processing is performed according to the contents of the stored priority table.
[0010]
Further, the computer according to the present invention is a computer that makes an access request to a plurality of storage devices that perform normal processing in response to an access request from the computer and a plurality of other background processing. A priority table indicating a processing priority between the normal processing and the plurality of background processing for each logical unit based on the determination result for each storage device. And a means for issuing them to each storage device.
[0011]
The storage device according to the present invention is a storage device that performs normal processing and a plurality of other background processing, and determines the use of each logical unit in the storage device based on an information table, and the determination A priority table indicating a processing priority between the normal processing and the plurality of background processing for each logical unit based on the result is provided, and processing is performed according to the contents of the priority table.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0013]
(First embodiment)
First, the first embodiment will be described.
FIG. 1 is a block diagram showing the configuration of a system to which the processing optimization method according to the first embodiment of the present invention is applied.
[0014]
In the figure, a plurality of host computers 10 to 1M (hereinafter referred to as hosts # 1 to #M) are provided, and these are connected to a network / channel network 30. The network / channel network 30 is connected to disk array devices 20 to 2N (hereinafter referred to as disk arrays # 0 to #N), which are a plurality of storage devices (storage).
[0015]
Hosts # 1 to #M issue an access request to an arbitrary disk array. On the other hand, the disk arrays # 0 to #N perform normal processing according to access requests from the hosts # 1 to #M and a plurality of other background processing (data transfer processing within / between cases). .
[0016]
Each of the hosts # 1 to #M includes an OS (operating system) 41, application software 42 (hereinafter referred to as an application 42), and a priority setting command issuing unit 43.
[0017]
The OS 41 receives a request from the application 42 and issues an access request to the corresponding disk array via the network / channel network 30. Further, the OS 41 sends a command issued from the priority setting command issuing unit 43 to the network / channel network 30.
[0018]
The application 42 makes a request for reading / writing data to / from the disk arrays # 0 to #N.
[0019]
The priority setting command issuing unit 43 sets the processing priority between the normal processing and the plurality of background processing for each logical unit according to the use of each logical unit (LU: used as a volume) in each disk array. A priority table is created for each disk array and issued to each disk array.
[0020]
In each of the disk arrays # 0 to #N, in addition to a plurality of disks (for example, HDDs) constituting the logical units LU # 0 to LU # m, an internal bus 51, a host interface 52, a microprocessor 53, and a disk side Control software 54 is provided.
[0021]
The internal bus 51 connects a plurality of disks (for example, HDDs), a microprocessor 53, and a host interface 52 that constitute the logical units LU # 0 to LU # m in the disk array.
[0022]
The host interface 52 performs an interface between the internal bus 51 and the hosts # 1 to #M.
[0023]
The microprocessor 53 controls the operation of the entire disk array apparatus and executes disk-side control software.
[0024]
The disk-side control software 54 acquires and stores the priority table issued from the hosts # 1 to #M, and performs processing according to the contents of the stored priority table.
[0025]
FIG. 2 is a block diagram illustrating a configuration of main parts of the system according to the first embodiment. In addition, the same code | symbol is attached | subjected to the element which is common in FIG.
[0026]
The priority setting command issuing unit 43 in the host 10 includes an information analyzing unit 45 and a command issuing unit 46.
[0027]
The information analysis unit 45 determines the usage of each logical unit in each disk array in the entire system based on the information table as shown in FIG. 3, and based on the determination result, normal processing and a plurality of logical units for each logical unit are determined. The processing priority during the background processing is determined for each disk array.
[0028]
In the example of the information table in FIG. 3, regarding the disk array # 0, the usage of LU # 0 is “active (normal operation)” and the usage of LU # 1 is “temporary disk”. It is shown. In this case, the priority within the same chassis is “1” for LU # 0 and “2” for LU # 1. Also for disk array # 1, the usage and priority within the same housing are shown for each logical unit, such as the usage of LU # 0 being "standby system (backup disk)".
[0029]
Based on the result of the information analysis using the information table, the information analysis unit 45 creates a priority table indicating the processing priority between the normal process and the plurality of background processes for each logical unit for each disk array. .
[0030]
The command issuing unit 46 adds a predetermined command (instruction to store the priority table) to the priority table created for each disk array by the information analysis unit 45 and issues it to each disk array. To do.
[0031]
On the other hand, the disk-side control software 54 in each disk array is provided with a priority control unit 55 and a priority table 56. The disk-side control software 54 acquires the priority table 56 sent to the corresponding disk array via the network / channel network 30, and stores it in a predetermined storage area.
[0032]
The priority control unit 55 controls normal processing and a plurality of background processing according to the contents of the stored priority table 56.
[0033]
An example of the priority table 56 is shown in FIG. As can be seen from FIG. 8, for LU # 0, “normal operation” is set to “1”, which is the highest priority, and “disk repair processing”, “data copy within disk unit”, “data between disk units” A priority “2” is set for each of “copy”, “primary backup disk”, and “internal data check”. For LU # 1, “1”, which is the highest priority, is set for “normal operation”, priority “2” is set for “disk repair processing”, and “data copy within disk device” is set. Is set to priority “3”, “copy data between disk devices” is set to priority “4”, “primary backup disk” is set to priority “5”, and “internal data check” is set. Has a priority “2”. Thus, processing priority is defined for each logical unit.
[0034]
Next, the operation according to the first embodiment will be described with reference to the flowchart of FIG.
The information analysis unit 45 in the host 10 determines the usage of each logical unit in each disk array in the entire system based on the information table, and performs normal processing and a plurality of backgrounds for each logical unit based on the determination result. Processing priority during processing is determined for each disk array (step A1). As a result, a priority table is created for each disk array.
[0035]
The command issuing unit 46 adds a predetermined command (instruction to store the priority table) to the created priority table and issues it to each disk array (step A2).
[0036]
When each disk array receives a command and priority table for itself, the disk control software 54 stores the priority table in a predetermined storage area according to the command (step A3).
[0037]
The priority control unit 55 controls the normal process and the plurality of background processes according to the contents of the stored priority table (step A4).
[0038]
Note that background processing other than normal processing (intra-housing / inter-housing data transfer processing) is instructed by a vendor-unique command (ModeSelect command in the case of SCSI). It is determined which data transfer instruction the processing is. Further, the control may be performed in consideration of the ratio of the number of I / Os, the occupation amount of the cache memory, and the like.
[0039]
As described above, according to the first embodiment, the priority indicating the processing priority between the normal processing and the plurality of background processing for each logical unit according to the use of each logical unit in each disk array. Since a table is created for each disk array and each process is performed based on this priority table, the process in each disk array can be optimized.
[0040]
(Second Embodiment)
Next, a second embodiment will be described.
Note that the configuration of the system to which the processing optimization method according to the second embodiment is applied is the same as that shown in FIG.
[0041]
FIG. 6 is a block diagram illustrating a configuration of main parts of a system according to the second embodiment. In addition, the same code | symbol is attached | subjected to the element which is common in FIG. In the following, the description will focus on the differences from the first embodiment (FIG. 2).
[0042]
In the above-described first embodiment (FIG. 2), the case where the disk-side control software 54 is provided with the priority control unit 55 and the priority table 56 has been described. In the second embodiment, information collection is additionally performed. An analysis unit 57 is provided.
[0043]
The information collection / analysis unit 57 collects information indicating various states in the disk array. For example, the following information is collected.
[0044]
-HDD statistics (such as frequent HDD retries)
-Information on each logical unit (such as a failure occurring in a RAID volume)
・ Current resource usage (cache memory, HDD, etc.)
Further, the information analysis unit 45 according to the present embodiment takes in the information collected by the information collection analysis unit 57 via the network / channel network 30. That is, the information analysis unit 45 considers not only the use of each logical unit but also the information acquired from the information collection analysis unit 57, and gives priority to processing between the normal processing and the plurality of background processing for each logical unit. The degree is determined for each disk array.
[0045]
Next, the operation according to the second embodiment will be described with reference to the flowchart of FIG.
The information analysis unit 45 in the host 10 takes out the information collected by the information collection analysis unit 57 of each disk array via the network / channel network 30 (step B1).
[0046]
Then, the information analysis unit 45 determines the usage of each logical unit in each disk array in the entire system based on the information table, determines various states indicated in the information acquired from the information collection analysis unit 57, and Based on these, the processing priority between the normal processing and the plurality of background processing for each logical unit is determined for each disk array (step B2). As a result, a priority table is created for each disk array.
[0047]
The command issuing unit 46 adds a predetermined command (instruction to store the priority table) to the created priority table, and issues it to each disk array (step B3).
[0048]
When each disk array receives a command and priority table for itself, the disk control software 54 stores the priority table in a predetermined storage area according to the command (step B4).
[0049]
The priority control unit 55 controls normal processing and a plurality of background processing in accordance with the contents of the stored priority table (step B5).
[0050]
Note that background processing other than normal processing (intra-housing / inter-housing data transfer processing) is instructed by a vendor-unique command (ModeSelect command in the case of SCSI). It is determined which data transfer instruction the processing is. Further, the control may be performed in consideration of the ratio of the number of I / Os, the occupation amount of the cache memory, and the like.
[0051]
The information analysis unit 45 according to the second embodiment takes in the information of the information collection analysis unit 57 constantly or periodically, and the contents of the priority table according to the change in the state in each disk array. Can be changed dynamically. Some specific examples are shown below.
[0052]
For example, if the HDD that constitutes LU # 0 of disk array # 0 has failed and there is no redundancy, the priority of normal access to logical units other than LU # 0 is lowered for a while to restore LU # 0. Prioritize.
[0053]
For example, LU # 0 (active system) of disk array # 0 and LU # 0 (standby system) of disk array # 1 are mirrored, and logical units other than LU # 0 of each disk array are used for different purposes. When the disk array # 0 stops in use, the priority is determined so that the priority of LU # 0 of the disk array # 1 is temporarily raised to function as an active system. After restoration of disk array # 0, priority is determined so that data copy from disk array # 1 to # 0 is given priority.
[0054]
Here, the operation of the case (1) will be described with reference to the flowchart of FIG.
[0055]
If a failure occurs in LU # 0 of disk array # 0 (step C1), whether the mirror volume of LU # 0 of disk array # 0 is in another chassis / in the same chassis (whether there is redundancy) ) Is determined (step C2).
[0056]
If there is a mirror volume, the process proceeds to step C6 and the disk is repaired. On the other hand, if there is no mirror volume, it is determined whether or not the priority of the disk repair process of LU # 0 of disk array # 0 is the highest setting (step C3).
[0057]
If the setting has the highest priority, the process proceeds to step C6 and the disk is repaired. On the other hand, if it is not the highest setting, the current priority is stored in a predetermined storage area, and the priority of the disk repair process of LU # 0 of disk array # 0 is set to the highest (step C5). Disk repair is performed (step C6).
[0058]
After the disk restoration is completed, the priority level saved in step C4 is restored (step C7).
[0059]
Next, the operation of the case (2) will be described with reference to the flowcharts of FIGS.
[0060]
When the disk array # 0 is stopped (step D1), it is determined whether the mirror volume other than the disk array # 0 is in another casing / in the same casing (step D2).
[0061]
If there is no mirror volume, the process proceeds to step D7 where disk array # 0 is repaired. On the other hand, if there is a mirror volume, by checking each logical unit in the disk array # 1, it is determined whether or not the priority of LU # 0 of the disk array # 1 has the highest setting ( Step D3).
[0062]
If the priority is set to the highest priority, the process proceeds to step D7 where disk array # 0 is repaired. On the other hand, if it is not the highest setting, it is determined whether or not there is no influence even if the priority of the disk array # 1 other than LU # 0 is lowered (step D4).
[0063]
If there is an influence, the process proceeds to step D7 where disk array # 0 is repaired. On the other hand, if there is no influence, the current priority is stored in a predetermined storage area (step D5), the operation priority of LU # 0 of disk array # 1 is set to the highest (step D6), and disk array # 0. Is restored.
[0064]
After the disk repair is completed, the priority of data transfer processing from LU # 0 of disk array # 1 to LU # 0 of disk array # 0 is set to the highest (step D8), and data copy is executed (step D8). D9), the priority saved in step D5 is restored (step D10).
[0065]
As described above, according to the second embodiment, the processing priority between the normal processing and the plurality of background processing for each logical unit depends on the use and various states of each logical unit in each disk array. Is created for each disk array, and each process is performed based on this priority table. Therefore, the optimization of the process can be further enhanced as compared with the first embodiment. .
[0066]
【The invention's effect】
As described above in detail, according to the present invention, in a system including a plurality of storage devices that perform normal processing in response to an access request from a computer and a plurality of other background processing, the processing of the entire system is optimized. Can be
[Brief description of the drawings]
FIG. 1 is a block diagram showing the configuration of a system to which a processing optimization method according to a first embodiment of the present invention is applied.
FIG. 2 is a block diagram showing a configuration of main parts of the system in the first embodiment.
FIG. 3 is a diagram showing an example of an information table used in the information analysis unit in the first embodiment.
FIG. 4 is a diagram showing an example of a priority table stored in the storage device in the first embodiment.
FIG. 5 is a flowchart for explaining the operation according to the first embodiment;
FIG. 6 is a block diagram showing the main configuration of a system according to a second embodiment of the present invention.
FIG. 7 is a flowchart for explaining an operation according to the second embodiment;
FIG. 8 is a flowchart for explaining a first example of dynamically changing a priority table in the second embodiment.
FIG. 9 is a flowchart for explaining a second example of dynamically changing the priority table in the second embodiment.
FIG. 10 is a flowchart for explaining a second example of dynamically changing the priority table in the second embodiment.
[Explanation of symbols]
10 to 1M Host computer (# 1 to #M)
20 to 2N: disk array device (# 1 to #N)
30 ... Network / channel network 41 ... OS (Operating System)
42 ... Application software 43 ... Priority setting command issuing unit 45 ... Information analyzing unit 46 ... Command issuing unit 51 ... Internal bus 52 ... Host interface 53 ... Microprocessor 54 ... Disk side control software 55 ... Priority control unit 56 ... Priority Table 57 ... Information collection and analysis section

Claims (9)

Processing applied to a system including a computer that issues an access request to an arbitrary storage device, and a plurality of storage devices that perform normal processing according to the access request from the computer and a plurality of other background processing An optimization method,
In the computer, the use of each logical unit in each storage device is determined based on the information table, and the processing priority between the normal processing and the plurality of background processing for each logical unit is determined based on the determination result. Create a priority table for each storage device and issue them to each storage device,
In each of the storage devices, the issued priority table is acquired and stored, and processing is performed according to the contents of the stored priority table.   The process optimization method according to claim 1, wherein the computer acquires information indicating a state in each storage device, and creates the priority table according to the acquired information and the application.   3. The process optimization method according to claim 2, wherein the contents of the priority table are dynamically changed according to a change in the state in each storage device. A computer that makes an access request to a plurality of storage devices that perform normal processing in response to an access request from a computer and a plurality of other background processing,
A priority table that determines the use of each logical unit in each storage device based on the information table and indicates the processing priority between the normal processing and the plurality of background processing for each logical unit based on the determination result A computer comprising: means for creating a storage device for each storage device and issuing them to each storage device.   5. The computer according to claim 4, wherein information indicating a state in each storage device is acquired, and the priority table is created according to the acquired information and the usage.   6. The computer according to claim 5, wherein the contents of the priority table are dynamically changed according to a change in the state in each storage device. A storage device that performs normal processing and a plurality of other background processing,
A priority table that determines the use of each logical unit in the storage device based on an information table and indicates the processing priority between the normal process and the plurality of background processes for each logical unit based on the determination result And a storage device that performs processing according to the contents of the priority table.   8. The storage device according to claim 7, further comprising means for collecting information indicating a state in the storage device, wherein the priority table is created according to the information and the application.   9. The storage device according to claim 8, wherein the content of the priority table is dynamically changed in accordance with a change in the state in the storage device.

Images