Multiplexing streams without changing the number of streams of a deduplicating storage system

Patent metadata

Field	Value
Publication number	US-9904480-B1
Application number	US-201414575079-A
Country	US
Kind code	B1
Filing date	Dec 18, 2014
Priority date	Dec 18, 2014
Publication date	Feb 27, 2018
Grant date	Feb 27, 2018

Abstract

In one embodiment, a method includes creating a first number of streams for a file system manager of a deduplicating storage system to access concurrently a type of data blocks, where each stream is for one file system and is identified by a stream identifier. The method further includes mapping stream identifiers to each of the type of data blocks passing through the first number of streams. The method further includes accessing the type of data blocks in storage units of the deduplicating storage system through a second number of streams, where the second number of streams are dedicated to the type of data blocks in the deduplicating storage system, where the second number is smaller than the first number, where the data blocks are tracked according to the mapped stream identifiers, and where the data blocks are stored in the storage units after a deduplication process to remove duplication.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
creating a first number of streams for a file system manager of a deduplicating storage system to access concurrently a type of data blocks, wherein each stream is for one file system and is identified by an identifier;
mapping stream identifiers to each of the type of data blocks passing through the first number of streams, wherein each data block is mapped to one stream identifier of one of the first number of streams that the data block passes through; and
accessing the type of data blocks in storage units of the deduplicating storage system through a second number of streams, wherein the second number of streams are preconfigured and dedicated to the type of data blocks for access in the deduplicating storage system, wherein the second number of streams is smaller than the first number of streams, wherein the data blocks are tracked according to the mapped stream identifiers, wherein the data blocks are stored in the storage units after a deduplication process to remove duplication, wherein the type of data block is file system namespace of the file system, wherein the data blocks of file system namespace are organized in a hierarchical data structure containing a layer of one root page, a layer of one or more intermediate pages, and a layer of a plurality of leaf pages, and wherein each leaf page contains one or more data blocks and each data block containing a namespace entry corresponding to a file or a directory of a file system of the deduplicating storage system.

2. The method of claim 1 , wherein mapping the stream identifiers is performed through a block map that maps each data block to one stream identifier.

3. The method of claim 1 , wherein the accessing the type of data blocks in storage units causes updating a bit map indicating a data block being occupied or not.

4. The method of claim 1 , further comprising:
updating snapshot status indications within the data blocks upon capturing a snapshot of the storage units of the deduplicating storage system.

5. The method of claim 1 , wherein the hierarchical data structure is a B+ tree data structure, and each intermediate page is an internal page of the B+ tree, and each internal page is a parent page of one or more leaf pages.

6. The method of claim 1 , wherein the tracking of the data blocks include indicating a page type of each data block, wherein the page type includes at least one of a summary page, a root page, an intermediate page, and a leaf page.

7. A deduplicating storage system, comprising:
a processor and a non-transitory computer readable storage medium coupled to the processor, the non-transitory machine-readable storage medium containing instructions executable by the processor, wherein the deduplicating storage system is operative to:
create a first number of streams for a file system manager of the deduplicating storage system to access concurrently a type of data blocks, wherein each stream is for one file system and is identified by an identifier;
map stream identifiers to each of the type of data blocks passing through the first number of streams, wherein each data block is mapped to one stream identifier of one of the first number of streams that the data block passes through; and
access the type of data blocks in storage units of the deduplicating storage system through a second number of streams, wherein the second number of streams are preconfigured and dedicated to the type of data blocks for access in the deduplicating storage system, wherein the second number of streams is smaller than the first number of streams, wherein the data blocks are tracked according to the mapped stream identifiers, wherein the data blocks are stored in the storage units after a deduplication process to remove duplication, wherein the type of data block is file system namespace of the file system, wherein the data blocks of file system namespace are organized in a hierarchical data structure containing a layer of one root page, a layer of one or more intermediate pages, and a layer of a plurality of leaf pages, and wherein each leaf page contains one or more data blocks and each data block containing a namespace entry corresponding to a file or a directory of a file system of the deduplicating storage system.

8. The deduplicating storage system of claim 7 , wherein mapping of the stream identifiers to the type of data blocks is to be performed through a block map that maps each data block to one stream identifier.

9. The deduplicating storage system of claim 7 , wherein the access of the type of data blocks in storage units is to cause updating a bit map indicating a data block being occupied or not.

10. The deduplicating storage system of claim 7 , the deduplicating storage system is further to:
update snapshot status indications within the data blocks upon capturing a snapshot of the storage units of the deduplicating storage system.

11. The deduplicating storage system of claim 7 , wherein the hierarchical data structure is a B+ tree data structure, and each intermediate page is an internal page of the B+ tree, and each internal page is a parent page of one or more leaf pages.

12. The deduplicating storage system of claim 7 , wherein the tracking of the data blocks include indicating a page type of each data block, wherein the page type includes at least one of a summary page, a root page, an intermediate page, and a leaf page.

13. A non-transitory computer-readable storage medium having instructions stored therein, which when executed by a processor, cause the processor to perform operations, the operations comprising:
creating a first number of streams for a file system manager of a deduplicating storage system to access concurrently a type of data blocks, wherein each stream is for one file system and is identified by an identifier;
mapping stream identifiers to each of the type of data blocks passing through the first number of streams, wherein each data block is mapped to one stream identifier of one of the first number of streams that the data block passes through; and
accessing the type of data blocks in storage units of the deduplicating storage system through a second number of streams, wherein the second number of streams are preconfigured and dedicated to the type of data blocks for access in the deduplicating storage system, wherein the second number of streams is smaller than the first number of streams, wherein the data blocks are tracked according to the mapped stream identifiers, wherein the data blocks are stored in the storage units after a deduplication process to remove duplication, wherein the type of data block is file system namespace of the file system, wherein the data blocks of file system namespace are organized in a hierarchical data structure containing a layer of one root page, a layer of one or more intermediate pages, and a layer of a plurality of leaf pages, and wherein each leaf page contains one or more data blocks and each data block containing a namespace entry corresponding to a file or a directory of a file system of the deduplicating storage system.

14. The non-transitory computer-readable medium of claim 13 , mapping the stream identifiers is performed through a block map that maps each data block to one stream identifier.

15. The non-transitory computer-readable medium of claim 13 , wherein the accessing the type of data blocks in storage units causes updating a bit map indicating a data block being occupied or not.

16. The non-transitory computer-readable medium of claim 13 , wherein the operations further comprise updatin

Assignees

• Emc Corp
• Emc Ip Holding Co Llc

Inventors

• Singh Pranay
• Chivukula Sai

Classifications

• G06F3/0619 Primary

  in relation to data integrity, e.g. data losses, bit errors · CPC title

• G06F3/0641

  De-duplication techniques · CPC title

• G06F3/0673

  Single storage device · CPC title

• G06F17/30088

  Physics · mapped topic

• G06F16/128

  Details of file system snapshots on the file-level, e.g. snapshot creation, administration, deletion (error detection or correction of the data by redundancy in operations or in hardware G06F11/14, G06F11/16) · CPC title