Erase block state detection

• US 10,108,355 B2
• Filed: 09/01/2015
• Issued: 10/23/2018
• Est. Priority Date: 09/01/2015
• Status: Active Grant

• Current Assignee: Pure Storage, Inc.
• Original Assignee: Pure Storage, Inc.
• Inventors: Hayes, John Martin, Kannan, Hari, Miladinovic, Nenad
• Primary Examiner(s): Chung, Phung M
• Application Number:
  US14/842,687
• Publication Number:
  US 20170060451A1
• Time in Patent Office: 1,148 Days
• CPC Class Codes: G06F 11/1068 in sector programmable memories, e.g. flash disk G06F11/1072 takes precedence
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0652 Erasing, e.g. deleting, data cleaning, moving of data to a wastebasket
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]

Abstract

A method for erasure detection in a storage cluster includes establishing a connection, via a network, of a storage unit to one of a plurality of storage nodes of a storage cluster and determining, for at least one page of a storage memory of the storage unit, that the at least one page is erased. The storage unit is one of a plurality of storage units configured to store user data in memory of the storage units in accordance with direction from the plurality of storage nodes. The method includes communicating from the storage unit to the one of the plurality of storage nodes that the at least one page is erased.

Claims

1. A method for erasure detection in a storage cluster, performed by the storage cluster, comprising:

establishing a connection, via a network, of a storage unit to one of a plurality of storage nodes of the storage cluster;

determining, for at least one page of a storage memory of the storage unit, that the at least one page is erased, wherein the storage unit is one of a plurality of storage units configured to store user data in memory of the storage units in accordance with direction from the plurality of storage nodes;

communicating from the storage unit to the one of the plurality of storage nodes that the at least one page is erased; and

bypassing error correction of the at least one page, responsive to determining that the at least one page is erased.

2. The method of claim 1, further comprising:

performing a search through pages of a block of the storage memory to determine a boundary between erased pages and pages written in order in the block, the search including repetitions of the determining as performed by the storage unit for corresponding pages in the search.

3. The method of claim 1, further comprising:

performing an error correction code operation on raw page data to produce one of an error or status indication relating to the raw page data, wherein the performing the error correction code operation is downstream of where erase detection logic of the storage unit accesses the raw page data.

4. The method of claim 1, wherein the determining includes performing a logic operation on the at least one page.

5. The method of claim 1, wherein the communicating further comprises:

communicating from the storage unit to the one of the plurality of storage nodes information regarding a boundary that separates erased pages from written pages in the storage memory of the storage unit.

6. The method of claim 1, wherein the determining comprises:

counting bits, bytes or words of the at least one page to a programmable threshold value.

7. A storage cluster with erasure detection, comprising:

a plurality of storage nodes, each of the plurality of storage nodes configured to have ownership of a portion of user data;

a plurality of storage units, each of the plurality of storage units having storage memory, each of the plurality of storage units configured to store user data as directed by the plurality of storage nodes; and

each of the plurality of storage units configured to establish a connection, via a network, to at least one of the plurality of storage nodes, each of the plurality of storage units configured to detect if there is an erased page in the storage memory, to identify the erased page, to the at least one of the plurality of storage nodes, and to bypass error correction of the erased page responsive to the erased page being identified.

8. The storage cluster of claim 7, wherein each of the plurality of storage units is configured to output information regarding a boundary between erased pages and non-erased pages.

9. The storage cluster of claim 7, wherein the at least one erased page is identified through a vector register or an address register.

10. The storage cluster of claim 7, comprising:

a register in each of the plurality of storage nodes that indicates erase state of each page of the storage memory based on output of an erase detection logic.

11. The storage cluster of claim 7, further comprising:

a control module that cycles reads of an interface to read each page of a block of the storage memory; and

each of the plurality of storage units having an output indicating erase state of pages of the block of the storage memory.

12. The storage cluster of claim 7, comprising:

each of the plurality of storage units having hardware logic to detect the erased page in the storage memory; and

each of the plurality of storage nodes having an output that indicates which pages of the storage memory are erased and which pages of the storage memory are not erased.

13. A storage cluster with page erasure detection, comprising:

a plurality of storage units, each of the plurality of storage units having a plurality of pages of storage memory;

a plurality of storage nodes, coupled together as a storage cluster, each of the plurality of storage nodes configured to direct storage of user data in the plurality of storage units;

each of the plurality of storage units having, as an output to one or more of the plurality of storage nodes, an indicator of erase state of at least one of the plurality of pages of storage memory; and

each of the plurality of storage units configured to establish a connection, via a network, to at least one of the plurality of storage nodes, configured to determine the erase state of the at least one of the plurality of pages, configured to communicate the erase state of the at least one of the plurality of pages to the at least one of the plurality of storage nodes via the indicator, and configured to bypass error correction of the at least one of the plurality of pages responsive to communication that the erase state indicates the at least one of the plurality of pages is erased.

14. The storage cluster of claim 13, further comprising:

a control module coupled to a data interface of the storage memory and erase detection logic and configured to determine a boundary between erased pages of a block and written pages in the block, wherein the indicator is based on the boundary.

15. The storage cluster of claim 13, further comprising:

each of the plurality of storage nodes having a control module coupled to a data interface of the storage memory and erase detection logic, the control module configured to access each of a plurality of pages of a block in the storage memory via the data interface, so that the erase detection logic determines whether each of the plurality of pages is erased; and

the indicator configured to identify which of the plurality of pages are erased and which of the plurality of pages are not erased.

16. The storage cluster of claim 13, further comprising:

an error correction code module coupled to a data interface of the storage memory and configured to produce, from raw page data, corrected data and an error or status indication, wherein the data interface, erase detection logic, the indicator and the error correction code module are included in a single integrated circuit having the indicator as an output.

17. The storage cluster of claim 13, further comprising:

erase detection logic in each of the plurality of storage units, the erase detection logic includes a logical gate, the erase detection logic does not include an arithmetic logic unit (ALU) of a processor; and

the indicator includes one of;

a bit latch, a vector register based on the erase detection logic, or an address register configured to hold an address of a boundary between erased pages and non-erased pages of the storage memory.