Method, device and computer readable storage media for managing storage system
Embodiments include a method performed by a computing device. The method includes (a) receiving a plurality of access requests to access data of a multilayered storage system; (b) in response to determining that a first access request can be served from a top data layer, executing the first access request by accessing the top data layer; (c) in response to determining that a second access request cannot be served from the top data layer, determining whether a current concurrency number of the top data layer is less than a permitted concurrency number (PCN) of the top data layer; and (d) in response to determining that the current concurrency number of the top data layer is not less than the PCN, waiting until the current concurrency number of the top data layer is less than the PCN and then executing the second access request by accessing another data layer below the top layer.
1. A method performed by a computing device, the method comprising:
receiving a plurality of access requests to access data of a multilayered storage system;
in response to determining that a first access request of the plurality can be served from a top data layer of the multilayered storage system, executing the first access request by accessing the top data layer;
in response to determining that a second access request of the plurality cannot be served from the top data layer, determining whether a current concurrency number of the top data layer is less than a permitted concurrency number (PCN) of the top data layer; and
in response to determining that the current concurrency number of the top data layer is not less than the PCN of the top data layer, waiting until the current concurrency number of the top data layer is less than the PCN of the top data layer and then executing the second access request by accessing another data layer of the multilayered storage system below the top layer.
2. The method of claim 1 wherein the top data layer is a volatile memory cache and the other data layer is an array of storage disks.
3. The method of claim 1 wherein the top data layer is a top-level volatile memory cache and the other data layer is a lower-level solid-state disk (SSD) cache above an even lower-level array of storage disks.
4. The method of claim 3 wherein executing the second access request by accessing the other data layer includes, in response to determining that the second access request of the plurality can be served from the lower-level SSD cache, executing the second access request by accessing the lower-level SSD cache.
5. The method of claim 3 wherein executing the second access request by accessing the other data layer includes, in response to determining that the second access request of the plurality cannot be served from the lower-level SSD cache, executing the second access request by accessing the array of storage disks.
6. The method of claim 3 wherein executing the second access request by accessing the other data layer includes:
in response to determining that the second access request of the plurality cannot be served from the lower-level SSD cache, executing the second access request by accessing the lower-level SSD cache, determining whether a current concurrency number of the lower-level SSD cache is less than a PCN of the lower-level SSD cache; and
in response to determining that the current concurrency number of the lower-level SSD cache is not less than the PCN of the lower-level SSD cache, waiting until the current concurrency number of the to lower-level SSD cache is less than the PCN of the lower-level SSD cache and then executing the second access request by accessing the array of storage disks.
7. The method of claim 6 wherein the method further comprises:
setting a PCN of the array of storage disks to be equal to a product of:
a number of storage disks in the array;
a fraction of the storage disks used for storing data rather than parity information;
a number of concurrent access requests permitted by a storage disk; and
a health value of the array;
setting the PCN of the lower-level SSD cache to be equal to a fraction of the PCN of the array; and
setting the PCN of the of the top data layer to be equal to the PCN of the array minus the PCN of the lower-level SSD cache.
8. The method of claim 1 wherein the method further comprises:
setting a PCN of the array of storage disks to be equal to a product of:
a number of storage disks in the array;
a fraction of the storage disks used for storing data rather than parity information;
a number of concurrent access requests permitted by a storage disk; and
a health value of the array; and
setting the PCN of the of the top data layer to be equal to a fraction of the PCN of the array.
9. The method of claim 1 wherein executing the second access request by accessing another data layer of the multilayered storage system below the top layer includes:
incrementing the current concurrency number of the top data layer upon beginning to access the other data layer; and
decrementing the current concurrency number of the top data layer upon finishing to access the other data layer.
10. An apparatus comprising processing circuitry coupled to memory configured to:
receive a plurality of access requests to access data of a multilayered storage system;
in response to determining that a first access request of the plurality can be served from a top data layer of the multilayered storage system, execute the first access request by accessing the top data layer;
in response to determining that a second access request of the plurality cannot be served from the top data layer, determine whether a current concurrency number of the top data layer is less than a permitted concurrency number (PCN) of the top data layer; and
in response to determining that the current concurrency number of the top data layer is not less than the PCN of the top data layer, wait until the current concurrency number of the top data layer is less than the PCN of the top data layer and then execute the second access request by accessing another data layer of the multilayered storage system below the top layer.
11. A computer program product comprising a non-transitory computer-readable storage medium having computer-readable program instructions thereon, which when executed by processing circuitry, cause the processing circuitry to:
receive a plurality of access requests to access data of a multilayered storage system;
in response to determining that a first access request of the plurality can be served from a top data layer of the multilayered storage system, execute the first access request by accessing the top data layer;
in response to determining that a second access request of the plurality cannot be served from the top data layer, determine whether a current concurrency number of the top data layer is less than a permitted concurrency number (PCN) of the top data layer; and
in response to determining that the current concurrency number of the top data layer is not less than the PCN of the top data layer, wait until the current concurrency number of the top data layer is less than the PCN of the top data layer and then execute the second access request by accessing another data layer of the multilayered storage system below the top layer.