Storage system for increased cache reliability VI reading redundant disks into cache

A storage system comprising:

• a plurality of disks storing data to be accessed by a computer and redundant data for recovering the data;
• a cache memory for temporarily storing the data;
• a controller for reading data from the plurality of disks for storage into the cache memory before transferring the data to the computer, wherein
• the controller performs a reading operation, in response to a read request coming from the computer, of reading data from the plurality of disks and reading the redundant data from a number of the plurality of disks and storing the read data and the read redundant data in the cache memory,
• wherein the number of the plurality of disks is subjected to reading of the redundant data and indicates a number of disks allowed to fail for data recovery.

A method of reading data in a storage system having a plurality of disks each storing data to be accessed by a computer and redundant data used for recovering the data, a cache memory for temporarily storing the data in the plurality of disks, and a controller for reading the data from the plurality of disks for storage into the cache memory before transferring the data to the computer, the method comprising:

• reading data from the plurality of disks and reading the redundant data from a number of the plurality of disks and storing the read data and the read redundant data in the cache memory, wherein the number of the plurality of disks is subjected to reading of the redundant data and indicates a number of disks allowed to fail for data recovery;
• and
• when the data requested by the read request is not stored in the cache memory, recovering the requested data using the redundant data stored in the cache memory, and storing the recovered requested data into the cache memory;
• and
• transferring the requested data stored in the cache memory to the computer.

A computer program stored on a computer readable medium to be executed in a storage system, the storage system having a plurality of disks each storing data to be accessed by a computer and redundant data used for recovering the data, a cache memory for temporarily storing the data in the plurality of disks and a controller for performing data reading from the plurality of disks for storage into the cache memory before transferring the data to the computer, wherein the computer program comprises:

• data reading code for, in response to a read request from the computer, reading data from the plurality of disks and reading the redundant data from a number of the plurality of disks and storing the read data and the read redundant data in the cache memory, wherein the number of the plurality of disks is subjected to reading of the redundant data and indicates a number of disks allowed to fail for data recovery;
• data recovery code for, when the data requested by the computer is not stored in the cache memory, recovering the requested data using the redundant data stored in the cache memory, and storing the recovered requested data to the cache memory;
• and
• data transfer code for transferring the requested data stored in the cache memory to the computer.

A storage system adapted to be coupled to a host computer, the storage comprising:

• a plurality of disks storing data to be accessed and redundant data for recovering the data if necessary;
• a cache memory for temporarily storing the data;
• a controller for performing data reading from the plurality of disks and storing the read data into the cache memory before transferring the read data to the computer;
• a device state management table for storing information that indicates an extent to which the plurality of disks are available for data reading, and
• a corresponding redundancy table for storing information that indicates how many of the disks are allowed to be in a failed condition based on the information in the device state management table;
• wherein, in response to a request from the computer, the controller reads data from the plurality of disks and reads the redundant data from a number of the plurality of disks and stores the read data and the read redundant data in the cache memory;
• wherein the number of the plurality of disks is subjected to reading of the redundant data and indicates a number of disks allowed to fail for data recovery;
• wherein the controller determines how many of the plurality of disks are available for data reading, stores an indication of that result to the device state management table, and based on the result determines how many of the disks are to be subjected to reading of the redundant data;
• and
• wherein based on the number of disks not available for data reading and the information stored in the corresponding redundancy table, the controller determines how many of the disks are to be used for reading of the redundant data.