JP2005301561A - Disk array system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a disk array device capable of constructing RAID configuration information based on the information on file characteristics and priority performance elements designated by an operator. <P>SOLUTION: This disk array device is provided with a means for constructing the RAID configuration information of the disk array device based on information to be acquired by a means for requesting an operator to input file characteristics through an input/output panel connected to a disk array device and a means for requesting the operator to input priority performance elements at the time of constructing the RAID configuration. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a disk array device, and more particularly to a disk array device that provides effective configuration information for creating RAID configuration information.

  In the storage device management method and management device disclosed in Japanese Patent Application Laid-Open No. 2003-216348, the device configuration and the performance of the configuration are displayed to present the device performance, and the device administrator selects the RAID configuration. The present invention provides a method and apparatus capable of performing accurate management when an advanced system administrator performs a large system.

  The auxiliary storage device disclosed in Japanese Patent Laid-Open No. 6-268570 has means for monitoring a command and taking data size statistics, and means for changing the stripe size when data of a specific size exceeds a prescribed ratio. Auxiliary storage devices are provided.

  Conventionally, array configuration information in this type of disk array device is constructed by specifying the physical disk to be used, the RAID type, and the stripe size. Although it is important for the performance of a RAID device to have a RAID configuration that matches the characteristics of the file, it is often difficult for a general operator to specify the configuration information optimal for the disk array device to be used and the file characteristics.

  An object of the present invention is to provide a disk array device capable of constructing RAID configuration information based on file characteristics specified by an operator and information on priority performance elements.

  The disk array device according to the present invention includes means for requesting an operator to input file characteristics and means for requesting input of a priority performance element via an input / output panel connected to the disk array device when constructing a RAID configuration of the disk array device. Means for constructing RAID configuration information based on the obtained information;

  In the present invention, the means for constructing the RAID configuration information determines the RAID type and stripe length from the file characteristic table and priority performance element table specified by the operator. And there exists an effect as described below.

  The first effect is that information relating to the structure of the disk array device is designated by specifying the capacity and physical disk number, so that the knowledge of the hardware structure specific to the disk array device is reduced when constructing the array configuration information. It can be done.

  The second effect is that, since the hardware of the disk array device generates the configuration information, the problem of selecting the configuration information related to the hardware structure can be incorporated into the configuration information construction means.

  FIG. 1 shows a disk array device as an embodiment of the present invention. In FIG. 1, the disk array device 1 has an interface 101 and the main control unit 11 is connected to the host system 2, the main control unit 11 is connected to the service processor 13 by connection 103, and the service processor 13 is connected to the input / output panel 3 by the interface 102. The main control unit 11 is connected to the hard disk drives HDD41, HDD42, HDD43, HDD44, HD45, and HDD46 through a connection 104. The RAID configuration information 12 is accessible from the main control unit 11.

  Fig. 2-1 shows the display contents that require the operator to input the file characteristics, including the capacity, physical disk, access type, and data length. Specify information about the file to be used in the planned value column in Figure 2-1. In Fig. 2-2, what is emphasized when creating configuration information is specified in% in the weight column. The conditions specify items such as disk usage efficiency (the ratio of actual data to the total capacity because redundant data is included), processing speed, and file reliability.

  Fig. 3-1 and Fig. 3-2 show the screens where the operator has finished entering the specified items on the input screens of Fig. 1 and Fig. 2; the capacity is 8GB, the physical disk is 0 and 1, and the access type In this example, a file having the characteristics of R (random) and a data length of 4 KB is desired to create configuration information with a weight of 20% disk usage efficiency, 50% processing speed and 30% reliability.

  4 shows the configuration information constructed by the process of FIG. 5 based on the file characteristics of FIG. 3-1 and the conditions of FIG. 3-2, assigning 2 as the logical unit (LUN) number and physical disks 0 and 1 An example in which 1 is set as the RAID type and a stripe length of 2 KB is determined will be described.

  FIG. 5 is a process for determining the RAID type and stripe length. Based on the file characteristic table specified in FIG. 3A and the condition table in FIG. 3B, the configuration information (RAID type and stripe length) shown in FIG. Is generated.

  Figure 6 shows the RAID type determination table, and column numbers 601 to 612 indicate the column numbers for explanation. A cell marked with-in the decision table is defined as ignoring the condition.

  FIG. 7 is a diagram showing a disk array device according to the second embodiment. History information 14 is added to the disk array device 1 in FIG. 1, and the data format recorded in the history information 14 is shown in FIG. Shown in

  In the history information 14, the access type (S; sequential / R; random) and the average value of the data length are recorded.

In FIG. 1, when the operator tries to construct the configuration information of the disk array device, if the construction of the configuration information is started from the input / output panel 3, the service processor 13 displays the file characteristic table of FIG. 2-1 and the condition of FIG. 2-2. Although the table is displayed, the right column of each table is blank, and the operator is requested to input. The operator inputs the contents of each item on the input / output panel. At this time, the matters concerning the structure of the disk array device are the capacity required for the file and the designation of the physical disk to be used, and the rest are the data length and access type which are the characteristics of the file. When the designations shown in FIGS. 2-1 and 2-2 are obtained, the service processor 13 executes the process shown in FIG. 5 as the configuration information construction process. In the process 502 of FIG. 5, the number of physical disks acquired in FIGS. 3-1 and 3-2, the access type and the use efficiency of the disk specified in 3-2, the processing speed, and the weight of the file reliability The top two items are selected, and the items except for the adopted RAID type column in the RAID type selection table of FIG. 6 are determined, and processing for determining the RAID type is performed in step 503. In the examples of Fig. 3-1 and Fig. 3-2, the weight of the items of processing speed and file reliability is high, so the number of physical disks is 2, column 606 in Fig. 6 is selected and determined as RAID1 type. The Since physical disks 0 and 1 are designated, they are assigned to 0 and 1 as they are. The stripe length is 2 KB so that 4 KB is distributed every 2 KB. For a LUN, an empty number is found and assigned automatically. In this way, the RAID configuration information of FIG. 4 is obtained. RAID determination table As for the other columns in FIG. 6, columns 601 to 612 are selected according to the conditions, and the respective RAID types are determined. In FIG. 6, since there is no appropriate type for the column with the decision type X, the operator is prompted to change the condition on the decision information screen of FIG. The contents of the configuration information of FIG. 4 determined in this way are displayed on the input / output panel, and the operator can confirm the configuration information determined by the hardware. This configuration information is sent to the configuration information table 12 and defines the subsequent operation of the disk array device 1 as a RAID.
As another embodiment of the present invention, the basic configuration is as described above, but the method of specifying file characteristics is further devised. The configuration is shown in FIG. In this figure, the disk array device 1 is added to the disk array device 1 of FIG. 1 with history information 14 using non-volatile storage capable of retaining information when the power of the disk array device is turned off. The main control 11 accumulates the read ratio (ratio of read commands to all commands) and the data length shown in FIG. 7 for each logical unit each time a write or read command is executed, calculates an average value of past data lengths, and records history information 14. Update. When the operator configures a new logical file, if there is a file used in the past with respect to the read ratio and data length of the file designation in FIG. The array configuration information is created by replacing the read ratio and the data length of the planned value 3-1 with numerical values obtained from the operation history information of FIG. Since other operations are the same as those in the first embodiment, description thereof will be omitted.

  As described above, in this embodiment, when there is a similar logical file, the file characteristic information is obtained from the history information of the file, so that the array configuration information suitable for the use situation can be obtained. It is done.

It is a block diagram of the form of the 1st example of the present invention. It is a file characteristic table and condition table for demonstrating this invention. It is a file characteristic table for describing the present invention, a condition table with specified values. It is the determined structure information table for demonstrating this invention. It is a flowchart describing the processing of the RAID configuration information generation means of the present invention. It is a RAID type determination table processed by the RAID configuration information generation means of the present invention. It is a block diagram of the form of the disk array apparatus of 2nd Example of this invention. It is an operation history table of the 2nd example of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Disk array apparatus 2 Host system 3 Input / output panel 11 Main control part 12 Configuration information 13 Service processor 14 History information 16 RAID information generation means 41 Hard disk drive (HDD)
42 Hard Disk Drive (HDD)
43 Hard Disk Drive (HDD)
44 Hard disk drive (HDD)
45 Hard disk drive (HDD)
46 Hard Disk Drive (HDD)

Claims (1)

When creating RAID configuration information in a disk array device consisting of multiple disk units and control units, RAID configuration information is generated from the information input by the means for inputting file characteristics and the means for inputting priority performance elements. A disk array device that generates and operates RAID configuration information by means for

Images