Method, apparatus and computer program product for managing storage system
Techniques manage a storage system. In such techniques, data types of data in multiple storage devices associated with the storage system are obtained. The multiple storage devices are divided into a first group of storage devices and a second group of storage devices according to the data types. A first processing ability of the first group of storage devices and a second processing ability of the second group of storage devices are determined separately. Data distribution is scheduled among the multiple storage devices on the basis of the first processing ability and the second processing ability. With such techniques, processing abilities of various storage devices in a resource pool may be balanced, so that workloads of various storage devices may be balanced and further overall response speed of the resource pool may be increased.
1. A method for managing a storage system, the method comprising:
obtaining data types of data in multiple storage devices associated with the storage system;
dividing the multiple storage devices into a first group of storage devices and a second group of storage devices according to the data types, the data types comprising a first type of data for managing operations of the storage system and a second type of data to be used by users of the storage system;
storing at least the first type of data in the first group of storage devices;
storing the second type of data in the second group of storage devices;
determining a first processing ability of the first group of storage devices based on at least the first type of data being stored in the first group of storage devices; and
determining a second processing ability of the second group of storage devices based on the second type of data being stored in the second group of storage devices, each of the first processing ability and the second processing ability representing a respective processing ability to serve data access requests from users; and
scheduling data distribution among the multiple storage devices on the basis of the first processing ability and the second processing ability.
2. The method of claim 1 , wherein:
the first group of storage devices stores data of both the first type and the second type.
3. The method of claim 2 , wherein scheduling data distribution among the multiple storage devices comprises:
moving data of the second type between the multiple storage devices.
4. The method of claim 3 , wherein moving data of the second type comprises:
determining respective workloads of the multiple storage devices; and
moving an extent for second-type data on a first storage device to a second storage device in response to determining a first workload of the first storage device among the multiple storage devices satisfies a predefined condition, the first workload being heavier than a second workload of the second storage device.
5. The method of claim 4 , wherein determining a first workload of a first storage device among the multiple storage devices satisfies a predefined condition comprises:
correcting the first workload and the second workload on the basis of the first processing ability and the second processing ability respectively; and
determining a first workload of a first storage device satisfies a predefined condition in response to determining the corrected first workload is heavier than the corrected second workload.
6. The method of claim 4 , wherein determining a first workload of a first storage device among the multiple storage devices satisfies a predefined condition comprises:
correcting respective workloads of the multiple storage devices on the basis of the first processing ability and the second processing ability; and
determining a first workload of a first storage device satisfies a predefined condition in response to determining the corrected first workload is higher than an average value of the respective workloads by a predefined ratio.
7. The method of claim 4 , wherein moving an extent for second-type data on the first storage device to the second storage device comprises:
selecting a source extent from multiple extents comprised in the first storage device, the source extent having a heavier workload than a further extent among the multiple extents; and
moving the source extent from the first storage device to a free extent on the second storage device.
8. The method of claim 7 , wherein the storage system is a Redundant Array of Independent Disks, and the method further comprises:
determining multiple extents comprised in a stripe associated with the source extent; and
selecting the second storage device from the multiple storage devices other than corresponding storage devices where the multiple extents reside.
9. The method of claim 4 , wherein determining respective workloads of the multiple storage devices comprises: for a given storage device among the multiple storage devices,
obtaining a historical workload of the given storage device;
determining a data type of data in the given storage device; and
correcting the historical workload with the first processing ability and the second processing ability according to the data type.
10. The method of claim 9 , wherein obtaining a historical workload of the given storage device comprises:
determining the historical workload of the given storage device on the basis of at least one of: a number of data access requests for the given storage device, the data amount involved in access to the given storage device, and response time of the given storage device.
11. The method of claim 1 , wherein the first type of data corresponds to system data, wherein the second type of data corresponds to user data, wherein storing at least the first type of data in the first group of storage devices includes storing at least the system data in the first group of storage devices, and wherein storing the second type of data in the second group of storage devices includes storing the user data in the second group of storage devices.
12. An apparatus for managing a storage system, comprising:
at least one processor;
a volatile memory; and
a memory coupled to the at least one processor and having instructions stored thereon, the instructions, when executed by the at least one processor, causing the apparatus to perform acts comprising:
obtaining data types of data in multiple storage devices associated with the storage system;
dividing the multiple storage devices into a first group of storage devices and a second group of storage devices according to the data types, the data types comprising a first type of data for managing operations of the storage system and a second type of data to be used by users of the storage system;
storing at least the first type of data in the first group of storage devices;
storing the second type of data in the second group of storage devices;
determining a first processing ability of the first group of storage devices based on at least the first type of data being stored in the first group of storage devices; and
determining a second processing ability of the second group of storage devices based on the second type of data being stored in the second group of storage devices, each of the first processing ability and the second processing ability representing a respective processing ability to serve data access requests from users; and
scheduling data distribution among the multiple storage devices on the basis of the first processing ability and the second processing ability.
13. The apparatus of claim 12 , wherein,
the first group of storage devices stores data of both the first type and the second type.
14. The apparatus of claim 13 , wherein scheduling data distribution among the multiple storage devices comprises:
moving data of the second type between the multiple storage devices.
15. The apparatus of claim 14 , wherein moving data of the second type comprises:
determining respective workloads of the multiple storage devices; and
moving an extent for second-type data on a first storage device to a second storage device in response to determining a first workload of the first storage device among the multiple storage devices satisfies a predefined condition, the first workload being heavier than a second workload of the second storage device.
16. The apparatus of claim 15 , wherein determining a first workload of a first storage device among the multiple storage devices satisfies a predefined condition comprises:
correcting the first workload and the second workload on the basis of the first processing ability and the second processing ability respectively; and
determining a first workload of a first storage device satisfies a predefined condition in response to determining the corrected first workload is heavier than the corrected second workload.
17. The apparatus of claim 15 , wherein determining a first workload of a first storage device among the multiple storage devices satisfies a predefined condition comprises:
correcting respective workloads of the multiple storage devices on the basis of the first processing ability and the second processing ability; and
determining a first workload of a first storage device satisfies a predefined condition in response to determining the corrected first workload is higher than an average value of the respective workloads by a predefined ratio.
18. The apparatus of claim 15 , wherein moving an extent for second-type data on the first storage device to the second storage device comprises:
selecting a source extent from multiple extents comprised in the first storage device, the source extent having a heavier workload than a further extent among the multiple extents; and
moving the source extent from the first storage device to a free extent on the second storage device.
19. The apparatus of claim 18 , wherein the storage system is a Redundant Array of Independent Disks, and the acts further comprise:
determining multiple extents comprised in a stripe associated with the source extent; and
selecting the second storage device from the multiple storage devices other than corresponding storage devices where the multiple extents reside.
20. The apparatus of claim 14 , wherein determining respective workloads of the multiple storage devices comprises: for a given storage device among the multiple storage devices,
obtaining a historical workload of the given storage device;
determining a data type of data in the given storage device; and
correcting the historical workload with the first processing ability and the second processing ability according to the data type.
21. The apparatus of claim 20 , wherein obtaining a historical workload of the given storage device comprises:
determining the historical workload of the given storage device on the basis of at least one of: a number of data access requests for the given storage device, the data amount involved in access to the given storage device, and response time of the given storage device.
22. A computer program product having a non-transitory computer readable medium which stores a set of instructions to manage a storage system; the set of instructions, when carried out by computerized circuitry, causing the computerized circuitry to perform a method of:
obtaining data types of data in multiple storage devices associated with the storage system;
dividing the multiple storage devices into a first group of storage devices and a second group of storage devices according to the data types, the data types comprising a first type of data for managing operations of the storage system and a second type of data to be used by users of the storage system;
storing at least the first type of data in the first group of storage devices;
storing the second type of data in the second group of storage devices;
determining a first processing ability of the first group of storage devices based on at least the first type of data being stored in the first group of storage devices; and
determining a second processing ability of the second group of storage devices based on the second type of data being stored in the second group of storage devices, each of the first processing ability and the second processing ability representing a respective processing ability to serve data access requests from users; and
scheduling data distribution among the multiple storage devices on the basis of the first processing ability and the second processing ability.