Storage system and storage administration method

1. A storage system comprising:

a processor; and

a plurality of physical storage drives that are each respectively housed in one of a plurality of drive enclosures,

wherein the processor

configures a virtual chunk with k (k is an integer equal to or larger than two) virtual parcels including a virtual parcel having user data and a respective virtual parcel having element data that is redundant data for repairing the user data, and

stores the virtual parcels included in a same virtual chunk by mapping the virtual parcels in storage areas of k of mutually different physical storage drives among N (k<N) of the plurality of physical storage drives,

maps the virtual parcels such that a maximum value of a numbers of the virtual parcels mapped to particular physical storage drive housed in a same drive enclosure among the plurality of drive enclosures is equal to or smaller than p/r, where redundancy of the virtual chunk is p, and a number of drive enclosures among the plurality of drive enclosures that are able to be active even if a failure occurs is r (r≤p).

2. The storage system according to claim 1 , wherein the processor further:

maps the virtual parcels such that numbers of the virtual parcels mapped to the particular physical storage drive to be housed in the same drive enclosure among the plurality of drive enclosures are a single number.

3. The storage system according to claim 1 , wherein

a number of pieces of the user data and a number of pieces of the redundant data that are included in a single virtual chunk,

a maximum number of the drive enclosures that includes a number of drive enclosures scheduled to be added, and

failure tolerance of the drive enclosures are prespecified at a time of initial construction of the storage system.

4. The storage system according to claim 3 , wherein,

when the number of the pieces of the user data included in the single virtual chunk is d p, additional physical storage drives are added in units of (d+p)r/p physical storage drives.

5. The storage system according to claim 3 , wherein,

when the number of the pieces of the user data included in the single virtual chunk is d, physical storage drives are added such that a number of the plurality of physical storage drives is an integer multiple of (d+p)r/p.

6. The storage system according to claim 1 , wherein,

in a case in which a new virtual chunk is to be configured, the processor

selects respective virtual parcels, and moves mapping of the respective virtual parcels between the plurality of physical storage drives,

moves the respective virtual parcels included in the same virtual chunk to the k mutually different physical storage drives among the N (k<N) physical storage drives when the respective virtual parcels are moved between the plurality of physical storage drives, and

further moves the respective virtual parcels such that the maximum value of the numbers of the respective virtual parcels mapped to a particular physical storage drive to be housed in a same drive among the plurality of drive enclosures is equal to or smaller than a predetermined value.

7. The storage system according to claim 1 , wherein

virtual stripe rows include the user data and the element data that is redundant data for repairing the user data, the user data and the element data being stripes included in mutually separate virtual parcels in the same virtual chunk,

the virtual chunk includes B (B is a positive integer) virtual stripe rows, and

each virtual parcel includes B stripes belonging to mutually different virtual stripe rows.

8. The storage system according to claim 7 , wherein

a virtual parity group includes k virtual storage drives,

a first virtual storage drive among the k virtual storage drives, wherein the first virtual storage drive stores the virtual parcel, and the k virtual storage drives store the virtual chunk,

each of the k virtual storage drives is mapped to a respective storage drive of the plurality of physical storage drives,

a Vchunk cycle group includes E (E is a positive integer) virtual parity groups that are included in a Vchunk cycle, and

a plurality of the virtual parity groups are cyclically allocated to particular physical storage drives in the Vchunk cycle group.

9. A storage administration method, the method comprising

configuring, by a processor, a virtual chunk with k (k is an integer equal to or larger than two) virtual parcels including a virtual parcel having user data and a virtual parcel having element data that is redundant data for repairing user data;

storing, by the processor, the virtual parcels included in a same virtual chunk by mapping the virtual parcels in storage areas of k of mutually different physical storage drives among N (k<N) of a plurality of physical storage drives; and

mapping the virtual parcels such that a maximum value of a numbers of the virtual parcels mapped to particular physical storage drives housed in a same drive enclosure among a plurality of drive enclosures is equal to or smaller than p/r, where redundancy of the virtual chunk is p, and a number of drive enclosures among the plurality of drive enclosures that are able to be active even if a failure occurs is r (r≤p).