Non-blocking caching for data storage drives

Patent metadata

Field	Value
Publication number	US-10698815-B2
Application number	US-201514788190-A
Country	US
Kind code	B2
Filing date	Jun 30, 2015
Priority date	Jun 30, 2015
Publication date	Jun 30, 2020
Grant date	Jun 30, 2020

Abstract

To provide enhanced operation of data storage devices and systems, various systems, apparatuses, methods, and software are provided herein. In a first example, a data storage device accumulates write data into a cache storage region prior to committing into an archive storage region and maintains a data structure that tracks the write data in the cache storage region. Responsive to receiving first write data into the cache storage region, the data storage device establishes first tracking elements in the data structure for the first write data in the cache storage region. Responsive to receiving second write data directed to storage locations overlapping the first write data, the data storage device accepts the second write data into the cache storage region and establishes second tracking elements in the data structure for the second write data in the cache storage region without modifying the first tracking elements.

First claim

What is claimed is:

1. A data storage device, comprising:
storage media comprising a cache storage region and an archive storage region, wherein:
the cache storage region comprises a random write storage region; and
the archive storage region comprises a shingled magnetic recording (SMR) storage region; and

a storage control system configured to:
accumulate write data associated with multiple write operations into the cache storage region;
for each of the multiple write operations, establish, in a data structure stored in the cache storage region, tracking elements that track one or more locations in the SMR storage region to which a corresponding portion of the write data is to be written, wherein at least a first portion and a second portion of the write data are destined for a same location in the SMR storage region;
prior to committing the write data to the SMR storage region, select commit data for transfer into the SMR storage region using the tracking elements, wherein the at least first portion and the second portion of the write data are merged such that only a last-accumulated write data of the at least first portion and the second portion of the write data is selected; and
responsive to the write data partially committing,
service from the cache storage region a read operation directed to the write data, wherein:
a first portion of the commit data is selected from the cache storage region for servicing the read operation from the at least first portion of the write data and a second portion of the commit data is selected from the cache storage region for servicing the read operation from the second portion of the write data, and
the first and the second portions of the commit data are selected for servicing the read operation based on which write data was received last in time.

2. The data storage device of claim 1 , wherein the storage control system is further configured to:
perform a commit process that burst transfers the commit data from the cache storage region into the SMR storage region.

3. The data storage device of claim 2 , wherein the storage control system is further configured to:
accept the multiple write operations without blocking until the commit process is performed; and
provide a write completion indication to a host system for each of the multiple write operations upon accumulation of the corresponding write data into the cache storage region.

4. The data storage device of claim 2 , wherein the storage control system is further configured to:
responsive to completion of the commit process for at least the commit data, invalidate at least corresponding ones of the tracking elements.

5. The data storage device of claim 1 , wherein the storage control system is further configured to:
during a commit process that transfers the commit data from the cache storage region into the SMR storage region, receive further write data directed to further storage addresses at least partially overlapping with storage addresses of the commit data; and
accept the further write data into the cache storage region without being blocked by the commit data being burst transferred.

6. The data storage device of claim 1 , wherein each tracking element of the tracking elements comprises sequence information and storage address indicators.

7. The data storage device of claim 1 , wherein the storage control system is further configured to:
responsive to read operations received by the data storage device, select first data blocks of read data from the first write data in the cache storage region and select second data blocks of the read data from the second write data in the cache storage region.

8. The data storage device of claim 1 , wherein the SMR storage region and the random write storage region are included on rotating magnetic media of a hard disk drive comprising the storage media.

9. A method of operating a data storage device with storage media that includes a cache storage region and an archive storage region, wherein the cache storage region comprises a random write storage region of the storage media and the archive storage region comprises a shingled magnetic recording (SMR) storage region, the method comprising:
accumulating write data associated with multiple write operations from a host system into the cache storage region;
responsive to accumulating the write data, establishing, in a data structure stored in the cache storage region, tracking elements that track a plurality of SMR storage region locations to which corresponding portions of the write data are to be stored, wherein a first portion and a second portion of the write data are to be stored in a same SMR storage region location in the SMR storage region;
prior to committing the write data to the SMR storage region, selecting commit data for transfer into the plurality of SMR storage region locations in the SMR storage region based, at least in part, on the tracking elements, wherein the first portion and the second portion of the write data are merged such that only a last accumulated write data of the first portion and the second portion of the write data is selected for inclusion in the commit data; and
responsive to the write data partially committing,
servicing from the cache storage region a read operation directed to the write data, wherein:
a first portion of the commit data is selected from the cache storage region for servicing the read operation from the first portion of the write data and a second portion of the commit data is selected from the cache storage region for servicing the read operation from the second portion of the write data, and
the first and the second portions of the commit data are selected for servicing the read operation based on which write data was received last in time.

10. The method of claim 9 , further comprising:
performing a commit process that burst transfers the commit data from the cache storage region into the SMR storage region.

11. The method of claim 10 , further comprising:
accepting the multiple write operations without blocking; and
providing a write completion indication to the host system for each of the multiple write operations upon storage of the corresponding write data into the cache storage region.

12. The method of claim 10 , further comprising:
responsive to completion of the commit process for at least the commit data, invalidating at least corresponding ones of the tracking elements.

13. The method of claim 9 , further comprising:
during a commit process that transfers the commit data from the cache storage region into the SMR storage region, receiving further write data directed to further storage addresses at least partially overlapping with storage addresses of the commit data; and
accepting the further write data into the cache storage region without being blocked by the commit data being burst transferred.

14. The method of claim 9 , wherein the tracking elements comprise sequence information and storage address indicators.

15. The method of claim 9 , further comprising:
responsive to read operations received by the data storage device, selecting first data blocks of read data from the first set of write data in the cache storage region and selecting second data blocks of the read data from the second set of write data in the cache storage region.

16. The method of claim 9 , wherein the SMR storage region and the random write storage region are included on rotating magnetic media of a hard disk drive comprising the storage media.

17. A data storage assembly, comprising:
a pluralit

Assignees

• HGST Netherlands BV
• Western Digital Tech Inc

Inventors

• Hess Randall L
• Nash Berck E
• Reiser James M
• Roberson Randy L
• Stokes Kris B
• Yandell Jesse L

Classifications

• G06F2212/1032

  Reliability improvement, data loss prevention, degraded operation etc · CPC title

• 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

• G06F12/0866 Primary

  for peripheral storage systems, e.g. disk cache · CPC title