Storage system and storage management method
Provided is a storage system and a storage management method, aiming at reducing data movement amount necessary for using an expanded capacity in a distributed RAID. When only A (A is a positive integer) physical storage drives are added, a storage controller selects virtual parcels that are mapped to different physical storage drives among N physical storage drives and are included in different virtual chunks, changes an arrangement of the selected virtual parcels to the added A physical storage drives, and constitutes a new chunk based on unallocated virtual parcels selected from different physical storage drives among the (N+A) physical storage drives.
1. A storage system that reduces data movement when a capacity of the storage system is increased, the storage system comprising:
a plurality of first physical storage drives,
one or more processors that are communicatively coupled to the plurality of first physical storage drives, wherein the one or more processors are collectively configured to:
form a plurality of virtual chunks with K virtual parcels, (K is an integer of 2 or more), each of the K virtual parcels having user data or element data that is redundant data for repairing the user data,
allocate at least one virtual chunk out of the virtual chunks to a virtual volume provided to a host,
respectively perform a mapping of virtual parcels included in a same virtual chunk to different first physical storage drives, and
process a new virtual chunk in response to a number A of a second physical storage drives being added to the first physical storage drives to increase the capacity of the storage system, wherein A is an integer of 1 or more but less than K;
wherein the new virtual chunk is processed by:
selecting K-A existing virtual parcels from different virtual chunks that existed prior to adding the new virtual chunk,
updating mapping location of the selected existing virtual parcels from regions in the first physical storage drives already allocated to the selected existing virtual parcels, to regions in the second physical storage drives, and
forming the new virtual chunk with K additional virtual parcels by setting K additional virtual parcels to K unallocated regions collected both from K-A unallocated regions in the first physical storage drives in response to updating the mapping location of the selected existing virtual parcels and from A regions in the second physical storage drives.
2. The storage system according to claim 1 , wherein
once the new virtual chunk is provided to the virtual volume, an available capacity provided to the host is increased.
3. The storage system according to claim 2 , wherein;
a predetermined number of the virtual chunks forms a virtual chunk group,
a respective storage region on each of the plurality of first physical storage drives collectively forms a parcel cycle, wherein each respective storage region of the parcel cycle has a same predetermined capacity,
the mapping includes mapping the virtual chunk group and the parcel cycle to each other, and
a capacity of the virtual volume is sequentially increased by performing the mapping of the virtual parcels.
4. The storage system according to claim 3 , wherein
the parcel cycle is mapped by a common mapping pattern.
5. The storage system according to claim 1
the new virtual chunk is formed by updating the mapping so the new virtual chunk is associated with a region of the second physical storage drives.
6. The storage system according to claim 1 , wherein
A is 2 or more; and
the new virtual chunk uses all the A added physical storage drives.
7. The storage system according to claim 1 , wherein the one or more processors are further collectively configured to:
form a plurality of virtual parity groups from the K virtual parcels, wherein each of the virtual parity group includes c (c is a positive integer) virtual chunks, and
form a period group from among the plurality of virtual parity groups, where the period group include E (E is a positive integer) virtual parity groups that each have a parcel cycle, and
wherein the virtual parcels are periodically allocated among the first physical storage and the second physical storage based on the period group drives.
8. A storage management method for reducing data movement when a capacity of storage system is increased, the storage management method comprising:
forming a plurality of virtual chunks with K virtual parcels, (K is an integer of 2 or more), each of the K virtual parcels having user data or element data that is redundant data for repairing the user data,
allocating at least one virtual chunk out of the virtual chunks to a virtual volume provided to a host,
respectively performing a mapping of virtual parcels included in a same virtual chunk to a different first physical storage drive among first physical storage drives of the storage system,
detecting a number A of second physical storage drives being added to the first physical storage drives to increase the capacity of the storage system, wherein A is an integer of 1 or more but less than K, and
processing a new virtual chunk using the second physical storage drives, wherein the new virtual chunk is processed by:
selecting K-A existing virtual parcels from different virtual chunks that existed prior to adding the new virtual chunk,
updating mapping location of the selected existing virtual parcels from regions in the first physical storage drives already allocated to the selected existing virtual parcels, to regions in the second physical storage drives, and
forming the new virtual chunk with K additional virtual parcels by setting K additional virtual parcels to K unallocated regions collected both from K-A unallocated regions in the first physical storage drives in response to updating the mapping location of the selected existing virtual parcels and from A regions in the second physical storage drives.
9. A non-transitory computer readable storage medium storing instructions for reducing data movement when a capacity of storage system is increased, the instructions when executed by one or ore more processors cause the one or more processor to collectively execute a method comprising:
forming a plurality of virtual chunks with K virtual parcels, (K is an integer of 2 or more), each of the K virtual parcels having user data or element data that is redundant data for repairing the user data,
allocating at least one virtual chunk out of the virtual chunks to a virtual volume provided to a host,
respectively performing a mapping of virtual parcels included in a same virtual chunk to a different first physical storage drive among first physical storage drives of the storage system, and
processing a new virtual chunk in response to a number A of second physical storage drives being added to the first physical storage drives to increase the capacity of the storage system, wherein A is an integer of 1 or more but less than K
wherein the new virtual chunk is processed by:
selecting K-A existing virtual parcels from different virtual chunks that existed prior to adding the new virtual chunk,
updating mapping location of the selected existing virtual parcels from regions in the first physical storage drives already allocated to the selected existing virtual parcels, to regions in the second physical storage drives, and
forming the new virtual chunk with K additional virtual parcels by setting K additional virtual parcels to K unallocated regions collected both from K-A unallocated regions in the first physical storage drives in response to updating the mapping location of the selected existing virtual parcels and from A regions in the second physical storage drives.