Extent metadata update logging and checkpointing

• US 9,268,653 B2
• Filed: 01/17/2014
• Issued: 02/23/2016
• Est. Priority Date: 01/17/2014
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Kimmel, Jeffrey S., Rakitzis, T. Byron
• Primary Examiner(s): Saeed, Usmaan
• Assistant Examiner(s): Weinrich, Brian E.
• Application Number:
  US14/158,689
• Publication Number:
  US 20150205680A1
• Time in Patent Office: 767 Days
• Field of Search: 707/649, 711/103
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1407 Checkpointing the instruction stream
• CPC Class Codes: G06F 11/1417 Boot up procedures
• CPC Class Codes: G06F 11/1438 Restarting or rejuvenating
• CPC Class Codes: G06F 11/1464 for networked environments
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 12/0246 in block erasable memory, e.g. flash memory
• CPC Class Codes: G06F 16/128 Details of file system snapshots on the file-level, e.g. snapshot creation, administration, deletion use of snapshots for error detection or correction G06F11/14, G06F11/16
• CPC Class Codes: G06F 16/2228 Indexing structures
• CPC Class Codes: G06F 16/2255 Hash tables
• CPC Class Codes: G06F 2201/80 Database-specific techniques
• CPC Class Codes: G06F 2201/84 Using snapshots, i.e. a logical point-in-time copy of the data

Abstract

In one embodiment, a cluster uses an extent store layer and a set of hash tables having a plurality of slots embodying extent metadata that describe write data of one or more write requests organized into one or more extents. One or more non-volatile logs (NVLogs) are maintained in the cluster. The one or more NVLogs include an extent store layer log maintained by the extent store layer. The extent store layer log records changes to the set of hash tables as a plurality of log stream structures, where each log stream structure is associated with a hash table. One or more storage devices of the cluster are organized as a plurality of log streams, where each log stream is associated with a corresponding log stream structure of the extent store layer log.

Claims

1. A system comprising:

a central processing unit (CPU) adapted to execute a storage input/output (I/O) stack having an extent storage layer;

a memory coupled to the CPU and configured to store the extent store layer of the storage I/O stack, the memory further configured to store a set of hash tables having a plurality of slots embodying extent metadata that describe write data of one or more write requests organized into one or more extents;

one or more non-volatile logs (NVLogs) coupled to the CPU, the one or more NVLogs including an extent store layer log maintained by the extent store layer, the extent store layer log configured to record changes to the set of hash tables as a plurality of log stream structures, each log stream structure associated with a hash table of the set of hash tables such that changed slots of the hash table are recorded as entries in the log stream structure;

one or more solid state drives (SSDs) coupled to the CPU and organized as a plurality of log streams, each log stream associated with a corresponding log stream structure of the extent store layer log and including one or more log stream pages storing copies of corresponding entries of the log stream structure; and

the CPU configured to write the hash table to the one or more SSDs using a fuzzy checkpointing technique wherein contents of a checkpoint are configured for applying log entry and log stream page updates occurring during the checkpoint.

2. The system of claim 1 wherein the memory is further configured to store an in-core copy of the corresponding log stream structure such that the in-core copy of the corresponding log stream structure is written to the log stream as the one or more log stream pages.

3. The system of claim 1 wherein the corresponding log stream structure is implemented as a double buffer to accommodate writing to the one or more SSDs while recording the entries to the corresponding log stream structure.

4. The system of claim 1 wherein each log stream page includes a pointer indicating a reverse chronological order location of a previous log stream page on the one or more SSDs.

5. The system of claim 1 wherein the checkpoint is not point-in-time consistent.

6. A method comprising:

executing, by a node of a cluster, an extent store layer of a storage input/output (I/O) stack;

storing, in the cluster, a set of hash tables having a plurality of slots embodying extent metadata that describe write data of one or more write requests organized into one or more extents;

maintaining one or more non-volatile logs (NVLogs) in the cluster, the one or more NVLogs including an extent store layer log maintained by the extent store layer, the extent store layer log configured to record changes to the set of hash tables as a plurality of log stream structures, each log stream structure associated with a hash table of the set of hash tables such that changed slots of the hash table are recorded as entries in the log stream structure;

organizing one or more solid state drives (SSDs) of the cluster as a plurality of log streams, each log stream associated with a corresponding log stream structure of the extent store layer log and including one or more log stream pages storing copies of corresponding entries of the log stream structure; and

writing the hash table to the one or more SSDs using a fuzzy checkpointing technique wherein contents of a checkpoint are configured for applying log entry and log stream page updates occurring during the checkpoint.

7. The method of claim 6 further comprising:

storing an in-core copy of the corresponding log stream structure such that the in-core copy of the corresponding log stream structure is written to the log stream as the one or more log stream pages.

8. The method of claim 6 wherein the corresponding log stream structure is implemented as a double buffer to accommodate writing to the one or more SSDs while recording the entries to the corresponding log stream structure.

9. The method of claim 6 wherein each log stream page includes a pointer indicating a reverse chronological order location of a previous log stream page on the one or more SSDs.

10. The method of claim 6 wherein the checkpoint is not point-in-time consistent.

11. A non-transitory computer readable medium including program instructions for execution on one or more processors, the program instructions when executed operable to:

store a set of hash tables having a plurality of slots embodying extent metadata that describe write data of one or more write requests organized into one or more extents;

maintain one or more non-volatile logs (NVLogs), the one or more NVLogs including an extent store layer log maintained by an extent store layer of a storage input/output (I/O) stack, the extent store layer log configured to record changes to the set of hash tables as a plurality of log stream structures, each log stream structure associated with a hash table of the set of hash tables such that changed slots of the hash table are recorded as entries in the log stream structure;

organize one or more solid state drives (SSDs) as a plurality of log streams, each log stream associated with a corresponding log stream structure of the extent store layer log and including one or more log stream pages storing copies of corresponding entries of the log stream structure; and

write the hash table to the one or more SSDs using a fuzzy checkpointing technique wherein contents of a checkpoint are configured for applying log entry and log stream page updates occurring during the checkpoint.

12. The non-transitory computer readable medium of claim 11 wherein the program instructions when executed are further operable to:

store an in-core copy of the corresponding log stream structure such that the in-core copy of the corresponding log stream structure is written to the log stream as the one or more log stream pages.

13. The non-transitory computer readable medium of claim 11 wherein the corresponding log stream structure is implemented as a double buffer to accommodate writing to the one or more SSDs while recording the entries to the corresponding log stream structure.

14. The non-transitory computer readable medium of claim 11 wherein each log stream page includes a pointer indicating a reverse chronological order location of a previous log stream page on the one or more SSDs.