Method and apparatus for data copy
Embodiments of the present disclosure relate to methods and apparatuses for data copy. A method comprises receiving, from a client module in a first processing unit, a read request for reading data from a first storage region. The method further comprises, in response to determining from the read request that the data is to be copied, obtaining, from a cache, a first bitmap associated with the first storage region, the first bitmap indicating if the first storage region is filled with zero. The method further comprises, in response to determining from the first bitmap that the first storage region is filled with zero, sending, to the client module, an indication that the data is zero.
1. A method performed by a cache module of a data storage system for fulfilling a data copy request, the method comprising:
receiving, from a client module operating on processing circuitry of the data storage system, a read request for reading data from a source storage region, the read request being part of a data copy operation initiated by the client module in response to the data copy request;
looking up the first storage region in a first zero-fill bitmap of the cache module to determine whether the source storage region is entirely empty;
in response to determining that the source storage region is entirely empty, sending a zero-fill indication to the client module indicating that the source storage region has been determined to be entirely empty;
in response to sending the zero-fill indication to the client module, receiving a write request for filling a destination storage region with all zeroes; and
in response to receiving the write request, updating a second zero-fill bitmap of the cache module to indicate that the destination storage region is entirely empty.
2. The method of claim 1 wherein:
the read request includes a flag indicating that the data is to be copied; and
the method further comprises determining from the read request that the data is to be copied by:
obtaining the flag from the read request; and
determining, based on the flag, if the data is to be copied.
3. The method of claim 1 wherein the cache module includes a plurality of zero-fill bitmaps, each zero-fill bitmap of the cache module representing a respective logical storage disk.
4. The method of claim 3 wherein:
the source storage region is on a first logical storage disk, the first zero-fill bitmap representing the first logical storage disk;
the destination storage region is on a second logical storage disk distinct from the first logical storage disk, the second zero-fill bitmap representing the second logical storage disk; and
the first zero-fill bitmap is distinct from the second zero-fill bitmap.
5. The method of claim 3 wherein:
the source storage region and the destination storage region are both on a single logical storage disk;
the first zero-fill bitmap is the second zero-fill bitmap;
the source storage region is represented by a first portion of the first zero-fill bitmap; and
the destination storage region is represented by a second portion of the first zero-fill bitmap distinct from the first portion.
6. The method of claim 1 wherein sending the zero-fill indication to the client module includes refraining from preparing a buffer representing the source storage region in the cache module full of zeroes.
7. The method of claim 6 wherein sending the zero-fill indication to the client module further includes embedding the indication within a scatter-gather list that points to an address that does not include a prepared buffer full of zeroes.
8. The method of claim 1 wherein determining that the source storage region is entirely empty includes refraining from reading data from the source storage region.
9. The method of claim 1 wherein the method further comprises, after updating the second zero-fill bitmap, writing zeroes to a persistent storage disk of the data storage system based on the second zero-fill bitmap.
10. The method of claim 1 wherein the client module and the cache module are both implemented on a same storage processing unit of the data storage system.
11. The method of claim 1 wherein:
the client module is implemented on a first storage processing unit of the data storage system;
the cache module is implemented on a second storage processing unit of the data storage system distinct from the first storage processing unit; and
the first storage processing unit is communicatively coupled to the second storage processing unit via a communication management interface.
12. An apparatus comprising:
persistent storage; and
processing circuitry coupled to memory configured to operate as a client module and a cache module, the cache module being configured to fulfill a data copy request by:
receiving, from the client module, a read request for reading data from a source storage region, the read request being part of a data copy operation initiated by the client module in response to the data copy request;
looking up the first storage region in a first zero-fill bitmap of the cache module to determine whether the source storage region is entirely empty;
in response to determining that the source storage region is entirely empty, sending a zero-fill indication to the client module indicating that the source storage region has been determined to be entirely empty;
in response to sending the zero-fill indication to the client module, receiving a write request for filling a destination storage region with all zeroes; and
in response to receiving the write request, updating a second zero-fill bitmap of the cache module to indicate that the destination storage region is entirely empty.
13. The apparatus of claim 12 wherein:
the read request includes a flag indicating that the data is to be copied; and
the cache module is further configured to fulfill the data copy request by determining from the read request that the data is to be copied by:
obtaining the flag from the read request; and
determining, based on the flag, if the data is to be copied.
14. The apparatus of claim 12 wherein the cache module includes a plurality of zero-fill bitmaps, each zero-fill bitmap of the cache module representing a respective logical storage disk backed by the persistent storage.
15. The apparatus of claim 12 wherein sending the zero-fill indication to the client module includes refraining from preparing a buffer representing the source storage region in the cache module full of zeroes.
16. The apparatus of claim 12 wherein determining that the source storage region is entirely empty includes refraining from reading data from the source storage region.
17. The apparatus of claim 12 wherein the cache module is further configured to fulfill the data copy request by, after updating the second zero-fill bitmap, writing zeroes to the persistent storage based on the second zero-fill bitmap.
18. The apparatus of claim 12 wherein:
the apparatus further includes a plurality of storage processing units including a first storage processing unit and a second storage processing unit; and
the client module and the cache module are both implemented on the first storage processing unit of the data storage system.
19. The apparatus of claim 12 wherein:
the apparatus further includes a plurality of storage processing units including a first storage processing unit and a second storage processing unit distinct from the first storage processing unit; and
the client module is implemented on the first storage processing unit;
the cache module is implemented on the second storage processing unit; and
the first storage processing unit is communicatively coupled to the second storage processing unit via a communication management interface.
20. A computer program product comprising a non-transient computer-readable storage medium that tangibly stores a set of machine-executable instructions, which, when executed by a computing device, cause the computing device to operate a client module and a cache module, the cache module being configured to fulfill a data copy request by:
receiving, from the client module, a read request for reading data from a source storage region, the read request being part of a data copy operation initiated by the client module in response to the data copy request;
looking up the first storage region in a first zero-fill bitmap of the cache module to determine whether the source storage region is entirely empty;
in response to determining that the source storage region is entirely empty, sending a zero-fill indication to the client module indicating that the source storage region has been determined to be entirely empty;
in response to sending the zero-fill indication to the client module, receiving a write request for filling a destination storage region with all zeroes; and
in response to receiving the write request, updating a second zero-fill bitmap of the cache module to indicate that the destination storage region is entirely empty.