KR19990051729A - Structure of Raid System with Dual Array Controllers - Google Patents

Abstract

The present invention relates to the architecture of a Redundant Arrays of Inexpensive Disks (RAID) system. Failure of the array controller in a raid system can have catastrophic consequences that halt the overall system operation as well as the raid system. Therefore, in systems where availability of the system is important, the array controller of the raid system is dually configured, and the two array controllers share disk storage devices through a small computer system interface (SCSI) bus. Even if a controller fails, another array controller can replace the failed array controller. However, in order to improve the performance of a raid system, in a raid system using a delayed write method using a disk cache memory in an array controller, when a single array controller fails, the disk cache memory in the controller writes to the disk. It is impossible to recover dirty blocks that could not be found. In order to solve this problem, in the present invention, a structure of a raid system having a dual array controller capable of facilitating data backup of a disk cache memory by allowing data transfer through the channel having a separate SCSI channel between the two array controllers is provided. Is presented.

Description

Structure of Raid System with Dual Array Controllers

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to the structure of a Redundant Arrays of Inexpensive Disks (hereinafter referred to as RAID) system. The present invention relates to a structure of a RAID system having a dual array controller employing a separate Small Computer System Interface (SCSI) channel between two array controllers so that data can be easily backed up from the disk cache memory of the controller.

RAID system improves I / O characteristics by bringing together multiple independent disk storage devices into one, distributing data to each disk and enabling access to multiple disks at the same time. This makes it possible to implement mass storage by making it look like a disk. It also accepts auxiliary data such as disk duplication, Error Checking and Correction (ECC) codes, or parity data at the same time, providing a way to automatically recover data in the event of a disk failure in a RAID system. It can increase the availability of the system.

RAID systems are divided into various levels according to their configuration. According to a paper published by Patterson et al., RAID systems are divided into five types, that is, RAID levels 1-5.

First, RAID level 1 stores data on N disks and copies the same data on another N mirror disks. When writing data, data should always be stored on two disks, and during reading, data can be read by selecting the faster disk among the two disks. If one disk fails, service can continue with the other disk. RAID level 2 uses Hamming code to protect data, which costs less disk than RAID level 1 mirroring. RAID level 3 consists of adding one parity disk to a group of N data disks. When writing data, data is stored in each data disk by interleaving the data in bit or byte units, and the parity calculated by exclusive-OR for the data stored in each data disk is stored in the parity disk. . When reading, N disks must be accessed at the same time. If one disk fails, information can be recovered using parity data stored in the parity disk. RAID level 4, like RAID level 3, consists of N + 1 disks. N disks store data and the other disk stores parity. The difference with RAID level 3 is that data is distributed in blocks. Therefore, when writing data, access of one data disk and parity disk is required, and only one disk needs to be accessed for reading. When a disk fails, information can be recovered using the parity stored in the parity disk. Similar to RAID level 4, RAID level 5 stores data in block units. However, RAID level 5 is a technique of distributing and storing parity on each disk like data without fixing parity on one disk. Reading and writing data and recovering data in the event of a disk failure are the same as for RAID level 4. In addition, there is RAID level 0, which does not use auxiliary data but simply distributes the data, and RAID level 6 with P + Q redundancy using Reed-Solomon code. RAID level 6 is more available to the system than using parity because information can be recovered even if two disks fail at the same time. Most RAID systems currently support RAID levels 0, 1, 3, 5, and RAID levels 0/1, which are a combination of RAID levels 0 and 1, and use a type suitable for the user's application environment.

In such a RAID system, a failure of the array controller can cause a fatal event that causes not only the RAID system but also the entire system to stop. Therefore, in a system where availability of the system is important, the array controller of the RAID system is dually configured, and the two array controllers share disk storage devices through the SCSI bus, so that one array controller fails even if one array controller fails. To replace a failed array controller.

1 is a structural diagram of a typical RAID system having a dual array controller, in which the first and second array controllers 12 and 13 are connected to a host computer 11, a SCSI bus and a disk storage block 14 for connecting disks. It is composed of

Disk storage devices 14 connected to the SCSI bus are divided into two logical groups by the initial configuration of the RAID system, each group belonging to one array controller 12 or 13. Thus, when both array controllers 12 and 13 are operating normally, each array controller controls only that disk group, and if one array controller 12 or 13 fails, the other array controller 13 or 12 ) Takes over control of the disk group belonging to the failed array controller 12 or 13.

However, in order to improve performance of a RAID system, in a RAID system using a delayed write method with a disk cache memory in an array controller, when a single array controller fails, the disk cache memory in the controller does not write to the disk. Recovery of bad dirty blocks is impossible.

Accordingly, the present invention provides a structure of a RAID system having a dual array controller that can facilitate data backup of the disk cache memory by allowing data transfer through this channel with a separate SCSI channel between the two array controllers. There is a purpose.

The structure of the raid system having a dual array controller according to the present invention for achieving the above object is a disk storage block consisting of a plurality of disks for information storage, and connected to the host computer via a host connection bus and fast data storage And a dual array controller comprising first and second array controllers having disk cache memory for writing and an inter-disk small computer system connection bus for connecting between said plurality of disks. And an array control period small computer system connection bus for connection between a controller and said second array controller.

1 is a structural diagram of a typical raid system.

2 is a structural diagram of a raid system according to the present invention.

3 is an array controller architecture diagram of a raid system according to the present invention;

4 is a functional diagram illustrating a method of operating a disk cache memory in a raid system according to the present invention.

<Description of the symbols for the main parts of the drawings>

21: host computer 22, 23: first and second array controller

24: Disk Storage Block 25: Array Control Period SCSI Bus

31: Processor 32: Local Memory

33: PCI bridge 34: disk cache memory

35 Parity Engine 36 Disk-to-Disk SCSI Controller Block

37: Array Control Period SCSI Controller

38: host interface unit

Hereinafter, with reference to the accompanying drawings will be described in detail the present invention.

2 is a structural diagram of a RAID system according to the present invention, in which a first array controller 22 and a second array controller 23 are connected to a host computer 21 through a host connection bus, and a disk storage block 24 Disks are connected via an interdisk SCSI bus. The first array controller 22 and the second array controller 23 are also connected via the SCSI bus 25 during the array control period.

Disk storage devices 24 connected to the inter-disk SCSI bus are divided into two logical groups by the initial configuration of the RAID system, and each group belongs to one array controller 22 or 23. Therefore, when both array controllers 22 and 23 operate normally, each array controller 22 and 23 controls only the corresponding disk group. If one array controller 22 or 23 fails, the other array The controller 23 or 22 takes over control of the disk group belonging to the failed array controller 22 or 23.

3 is a schematic diagram of an array controller of a RAID system according to the present invention.

As shown, an array controller in a RAID system includes a processor 31, a local memory 32, and a Peripheral Component Interconnect (PCI) bus that perform array management software and manage data transfer between a host computer and SCSI disk storage. It has a host interface 38 for data transfer with the same input / output bus and host computer. In addition, the SCSI controller block 36 of the array controller of the RAID system is connected to the disk storage device through the inter-disk SCSI bus, and the array control period SCSI controller 37 is connected to another array in the RAID system through the array control period SCSI bus. It is connected to the controller. In addition, the buffer cache memory 34 of the array controller is used as a disk cache or speed matching buffer, and the parity engine 35 is used for parity calculation. The PCI input / output bus is also connected to the processor 31 and the local memory 32 through the PCI bridge 33.

4 is a functional diagram illustrating a method of operating a disk cache memory in a RAID system according to the present invention.

In a RAID system with a dual array controller, the disk cache memory (34 in FIG. 3) in the array controller is divided into two blocks, one block serving as a normal cache memory for storing its data and the other block the other array controller. Used to back up disk cache memory. That is, the disk cache memory of the first array controller is divided into a normal first disk cache memory 41 and a cache memory 42 for backing up the second disk cache memory, and the disk cache memory of the second array controller is a normal second disk. The cache memory 43 is divided into a cache memory 44 for the first disk cache memory backup.

When a disk read request is generated from the host computer, the array controller checks whether the requested data block exists in the disk cache memory, and reads data from the disk cache memory if the cache hit state and reads data from the disk if the cache fails. For the disk write request, delayed write is used to improve the performance of the RAID system. It does not store the data on the physical disk but stores the data only in cache memory and sends a write completion signal to the host computer immediately. This is to achieve the effect of reducing the disk write time. After sending a write complete signal, the array controller updates the cache table in disk cache memory to indicate that the new data block is a dirty block that is different from the contents of the physical disk. The array control period transfers new data blocks and cache tables to the disk cache memory of the other array controller through the communication SCSI channel.

When the array controller becomes idle or there is no spare disk cache memory, it transfers dirty blocks of cache memory to disk and updates the cache table. In this case, only the cache table is transferred to another array controller. Ensure that the disk cache memory of the array controller is always in the same state. If one array controller fails, the host computer that detects it attempts to access a disk group belonging to the failed array controller through the other array controller. A normal array controller manages two logical disk groups by separating them into logical unit numbers (LUNs) of SCSI commands. The disk cache memory used for backup of the failed array controller continues to be used as cache memory for the LUN.

As described above, data transfer between array controllers is possible by providing an array control period SCSI channel separate from an inter-disk SCSI channel as a special transmission channel between two array controllers in hardware for copying data between array controllers.

As described above, according to the present invention, a separate SCSI channel is provided between two array controllers to provide a data transmission path and to keep the disk cache memory the same so that the data of the disk cache memory can be used while using the delayed write method. There is an excellent effect that can facilitate backups.

Claims (2)

A disk storage block consisting of a plurality of disks for storing information, First and second array controllers connected to a host computer via a host connection bus and having disk cache memory for fast reading and writing of data; 10. A raid system having a dual array controller comprised of inter-disk small computer system connection buses for connecting between said plurality of disks. An array control period comprising a small computer system connection bus for connection between said first array controller and said second array controller. 2. The raid system of claim 1, wherein the disk cache memory of the first and second array controllers is divided into a disk cache memory block for data storage and a disk cache memory block for data backup on the opposite side. Structure.