Managing input/output operations for shingled magnetic recording in a storage system

Patent metadata

Field	Value
Publication number	US-10073774-B2
Application number	US-201715459866-A
Country	US
Kind code	B2
Filing date	Mar 15, 2017
Priority date	Apr 29, 2016
Publication date	Sep 11, 2018
Grant date	Sep 11, 2018

Abstract

A system and method for improving the management of data input and output (I/O) operations for Shingled Magnetic Recording (SMR) devices in a network storage system is disclosed. The storage system includes a storage controller that receives a series of write requests for data blocks to be written to non-sequential addresses within a pool of SMR devices. The storage controller writes the data blocks from the series of write requests to a corresponding sequence of data clusters allocated within a first data cache of the storage controller for a thinly provisioned volume of the pool of SMR devices. Upon determining that a current utilization of the first data cache's data storage capacity exceeds a threshold, the sequence of data clusters including the data blocks from the first data cache are transferred to sequential physical addresses within the SMR devices.

First claim

What is claimed is:

1. A method, comprising:
receiving, by a storage controller from one or more host devices, a series of write requests for data blocks to be written to corresponding addresses within a pool of Shingled Magnetic Recording (SMR) devices of a network storage system, each write request in the series specifying a corresponding virtual address within a thinly provisioned volume of the pool of SMR devices to which a data block is to be written, further wherein the data blocks in the series of write requests are directed to non-sequential virtual addresses within the thinly provisioned volume of the pool of SMR devices;
in response to the series of write requests, writing the data blocks from the series of write requests to a corresponding sequence of data clusters allocated within a first data cache of the storage controller for the thinly provisioned volume of the pool of SMR devices;
mapping the non-sequential virtual addresses of the data blocks from the series of write requests to sequential physical addresses within the pool of SMR devices; and
upon determining that a current utilization of a data storage capacity of the first data cache exceeds a threshold, transferring the sequence of data clusters including the data blocks from the first data cache to the sequential physical addresses within the pool of SMR devices.

2. The method of claim 1 , wherein the threshold for the current utilization of the first data cache's data storage capacity is based on an SMR zone size associated with the pool of SMR devices.

3. The method of claim 1 , wherein the first data cache is a dynamic random access memory (DRAM) of the storage controller.

4. The method of claim 1 , further comprising:
storing, within a second data cache, metadata relating to the data clusters allocated for the thinly provisioned volume within the first data cache of the storage controller.

5. The method of claim 4 , wherein the second data cache includes one or more SSD devices coupled to the storage controller.

6. The method of claim 5 , further comprising:
receiving, by the storage controller from the one or more host devices, data read requests for data blocks to be read from the pool of SMR devices;
reading the data blocks from the pool of SMR devices, based on the received read requests; and
using the one or more SSD devices of the second data cache as a SSD read cache to store the data blocks that are read from the pool of SMR devices.

7. The method of claim 4 , wherein the metadata includes an index mapping the corresponding addresses of the data blocks within the pool of SMR devices to locations of the data blocks written to the sequence of data clusters allocated for the thinly provisioned volume within the first data cache of the storage controller.

8. The method of claim 4 , further comprising:
monitoring, by the storage controller, the current utilization of the first data cache's data storage capacity based on a total number of data clusters allocated within the first data cache for the thinly provisioned volume; and
storing, within the metadata, information related to the current utilization of the first data cache, based on the monitoring.

9. The method of claim 8 , wherein modifying the portion of the previously allocated data cluster comprises overwriting any existing content of the previously allocated data cluster with the other data block.

10. The method of claim 8 , wherein modifying the portion of the previously allocated data cluster comprises writing the other data block to the portion of the previously allocated data cluster such that the other data block overlays any existing content of the previously allocated data cluster within the first data cache.

11. The method of claim 1 , wherein each data cluster in the sequence of data clusters within the first data cache corresponds to a different address range within the pool of SMR devices, and the writing of the data blocks comprises:
determining whether or not a previously allocated data cluster within the first data cache exists for an address range that includes the corresponding virtual address specified for each of the data blocks;
in response to determining that the previously allocated data cluster exists within the first data cache for the address range including an address of a particular one of the data blocks, modifying at least a portion of the previously allocated data cluster within the first data cache to include the particular data block; and
in response to determining that the previously allocated data cluster does not exist within the first data cache for an address of another one of the data blocks:
allocating a new data cluster to which the other data block is to be written within the first data cache;
writing the other data block to the new data cluster allocated within the first data cache;
mapping the other data block's address within the first data cache to the corresponding virtual address within the thinly provisioned volume of the pool of SMR devices; and
storing metadata including information relating to the mapping of the other data block's address within the first data cache and the current utilization of the first data cache's data storage capacity, based on the allocation of the new data cluster and the writing of the other data block thereto.

12. A non-transitory machine readable medium having stored thereon instructions for performing a method comprising machine executable code which when executed by at least one machine, causes the machine to:
receive, from one or more host devices, a series of write requests for data blocks to be written to corresponding addresses within a pool of Shingled Magnetic Recording (SMR) devices of a network storage system, each write request in the series specifying a corresponding virtual address within a thinly provisioned volume of the pool of SMR devices to which a data block is to be written, further wherein the data blocks in the series of write requests are directed to non-sequential virtual addresses within the thinly provisioned volume of the pool of SMR devices;
write the data blocks from the series of write requests to a corresponding sequence of data clusters allocated within a first data cache of the storage controller for the thinly provisioned volume of the pool of SMR devices;
map the non-sequential virtual addresses of the data blocks from the series of write requests to sequential physical addresses within the pool of SMR devices;
monitor a current utilization of a data storage capacity of the first data cache; and
transfer the sequence of data clusters including the data blocks from the first data cache to the sequential physical addresses within the pool of SMR devices, when the current utilization is determined to exceed a threshold, based on the monitoring.

13. The non-transitory machine readable medium of claim 12 , wherein the threshold for the current utilization of the first data cache's data storage capacity is based on an SMR zone size associated with the pool of SMR devices.

14. The non-transitory machine readable medium of claim 12 , further comprising machine executable code that causes the machine to:
store, within a second data cache, metadata relating to the data clusters allocated for the thinly provisioned volume within the first data cache.

15. The non-transitory machine readable medium of claim 14 , wherein the metadata includes an index mapping the corresponding addresses of the data blocks within the pool of SMR devices to the locations of the data blocks written to the sequence of data clusters allocated for the thinly provisioned volume with

Assignees

• Netapp Inc

Inventors

• Jibbe Mahmoud K
• Holt Keith
• Terrill Scott

Classifications

• G06F12/0802 Primary

  Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches · CPC title

• G11B5/012 Primary

  Recording on, or reproducing or erasing from, magnetic disks (G11B17/00, G11B19/00 take precedence) · CPC title

• G06F12/10

  Address translation · CPC title

• G11B2020/10675

  aspects of buffer control · CPC title

• G06F3/067

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title