Storage system and method for backup

A storage system comprising:

• a disk controller which has a CPU, a main memory, and an interface;
• and
• a disk device which has original volumes for backup and a storage pool for backup data,
• wherein the main memory incorporates a differential management program which checks whether the original volumes for backup are updated or not, a pool management program which allocates a disk area for storage of backup data to the storage pool for backup data, a performance management program which manages the performance of each volume of the disk device, and a backup control program which issues an instruction to the differential management program, the pool management program, and the performance management program for total backup control,
• wherein the backup control program selects a backup method by which recovery within a user-specified recovery object time is possible, according to a restore performance calculated by the performance management program and the total size of changed blocks after backup acquisition, as counted by the differential management program,
• wherein full backup, differential backup, or incremental backup is selected as the backup method,
• wherein the restore performance is estimated based on a write performance or a read performance, whichever is lower;
• and
• wherein under the backup control program:
• an estimated restore time for incremental backup is calculated by dividing the cumulative total size of changed blocks after full backup by the estimated restore performance,
• a decision is made as to whether the calculated estimated restore time is within the recovery object time,
• if it is not within the recovery object time, the pool management program reallocates an area required for differential backup according to the total size of changed blocks as a difference from full backup and recalculates the estimated restore time for differential backup, and
• if the recalculated estimated restore time is within the object time, a changed part of the original volume for backup is copied into the allocated area in the storage pool for backup data to acquire a differential backup.

A storage system comprising:

• a disk controller which has a CPU, a main memory, and an interface;
• and
• a disk device which has original volumes for backup and a storage pool for backup data,
• wherein the main memory incorporates a differential management program which checks whether the original volumes for backup are updated or not, a pool management program which allocates a disk area for storage of backup data to the storage pool for backup data, a performance management program which manages the performance of each volume of the disk device, and a backup control program which issues an instruction to the differential management program, the pool management program, and the performance management program for total backup control,
• wherein the backup control program selects a backup method by which recovery within a user-specified recovery object time is possible, according to a restore performance calculated by the performance management program and the total size of changed blocks after backup acquisition, as counted by the differential management program,
• wherein full backup, differential backup, or incremental backup is selected as the backup method,
• wherein the restore performance is estimated based on a write performance or a read performance, whichever is lower, and
• wherein under the backup control program:
• an estimated restore time for incremental backup is calculated by dividing the total size of changed blocks after full backup by the estimated restore performance,
• a decision is made as to whether the calculated estimated restore time is within the recovery object time or not,
• if it is not within the recovery object time, the pool management program reallocates an area required for differential backup according to the total size of changed blocks as a difference from full backup and recalculates the estimated restore time for differential backup, and
• if the recalculated estimated restore time is within the object time, a full backup is made.

A storage system comprising:

• a disk controller which has a CPU, a main memory, and an interface;
• and
• a disk device which has original volumes for backup,
• wherein the storage system is connected through a data transfer line with a backup storage system which has a volume for storage of backup as a backup destination,
• wherein the main memory incorporates;
• a differential management program which checks whether the original volumes for backup are updated or not, a performance management program which manages the performance of each volume of the disk device, a data transfer program which transfers data between the storage system and the backup storage system, a backup destination management program which manages the backup volume for backup data, and a backup control program which issues an instruction to the differential management program, the performance management program, the data transfer program, and the backup destination management program for total backup control,
• wherein the backup control program selects a backup method by which recovery within a user-specified recovery object time is possible, according to a restore performance calculated by the performance management program and the total size of changed blocks after backup acquisition as counted by the differential management program,
• wherein the performance management program maintains a performance management table which is used to manage disk and volume performance and calculate the restore performance,
• wherein the performance management table includes for each volume at least a Redundant Array of Inexpensive Disks (RAID) configuration which constitute said volume, information indicating a read performance of said volume, and information indicating a write performance of said volume, and
• wherein the restore performance is calculated based on at least the RAID configuration, read performance and write performance of each volume.

A storage system comprising:

• a disk controller which has a CPU, a main memory, and an interface;
• and
• a disk device which has original volumes for backup;
• wherein the storage system is connected through a data transfer line with a backup storage system which has a storage volume for backup data as a backup destination,
• wherein the storage system is connected with a backup server which has backup software to manage backup data on a file-by-file basis and a backup setup program,
• wherein the main memory incorporates a differential management program which checks whether the original volumes for backup are updated or not, a performance management program which manages the performance of each volume of the disk device, a data transfer program which transfers data between the storage system and the backup storage system, a backup destination management program which manages the storage volume for backup data, and a backup control program which issues an instruction to the differential management program, the performance management program, the data transfer program, and the backup destination management program for total backup control,
• wherein the backup control program selects a backup method by which it is possible to recover data within a recovery object time specified on a setup screen under the backup setup program of the backup server, according to a restore performance calculated by the performance management program and the total size of changed blocks after backup acquisition as counted by the differential management program
• wherein the performance management program maintains a performance management table which is used to manage disk and volume performance and calculate the restore performance,
• wherein the performance management table includes for each volume at least a Redundant Array of Inexpensive Disks (RAID) configuration which constitute said volume, information indicating a read performance of said volume, and information indicating a write performance of said volume, and
• wherein the restore performance is calculated based on at least the RAID configuration, read performance and write performance of each volume.

A backup method for a storage system which comprises a disk controller which has a CPU, a main memory, and an interface, and a disk device which has original volumes for backup and a storage pool for backup data, wherein the main memory incorporates a differential management program which checks whether the original volumes for backup are updated or not, a pool management program which allocates a disk area for storage of backup data to the storage pool for backup data, a performance management program which manages the performance of each volume of the disk device, and a backup control program which issues an instruction to the differential management program, the pool management program, and the performance management program for total backup control,

• the method comprising the steps of:
• the differential management program counting the total size of changed blocks after a previous backup acquisition;
• the performance management program calculating estimated restore time for backup by reading a write performance and a read performance and taking the lower performance as an estimated restore performance and dividing the total size of changed blocks by the estimated restore performance;
• deciding whether the calculated estimated restore time is within a user-specified recovery object time or not;
• and
• selecting a backup method by which recovery within the object time is possible,
• wherein the performance management program maintains a performance management table which is used to manage disk and volume performance and calculate the restore performance,
• wherein the performance management table includes for each volume at least a Redundant Array of Inexpensive Disks (RAID) configuration which constitute said volume, information indicating a read performance of said volume, and information indicating a write performance of said volume, and
• wherein the restore performance is calculated based on at least the RAID configuration, read performance and write performance of each volume.