Reducing fragmentation in compressed journal storage

Patent metadata

Field	Value
Publication number	US-9703795-B2
Application number	US-201314074172-A
Country	US
Kind code	B2
Filing date	Nov 7, 2013
Priority date	Jan 2, 2013
Publication date	Jul 11, 2017
Grant date	Jul 11, 2017

Abstract

While discharging a data chunk the chunk is compressed into a storage block. If the chunk is found to be too large to be able to be completely compressed into the storage block certain characteristics of the data chunk are examined to determine whether or not the data chunk should be split. If the data chunk should be split, a remaining portion of the data chunk is compressed to a storage block that is next in chronological order to the original storage block. If the data chunk should not be split, all of the data chunk is moved to the next chronological storage block while leaving any remaining space in the storage block as unused.

First claim

What is claimed is:

1. A method for reducing fragmentation in compressed journal storage by a processor device in a computing environment, the method comprising:
in the compressed journal storage in which a user write size of a data chunk is unknown before compression, during the compression of the data chunk of a plurality of various sized compressed data blocks into a fixed size storage block when emitting the data chunk, and when the data chunk is unable to be completely compressed into the storage block after attempting to completely compress the data chunk into the storage block:
analyzing a plurality of attributes of the data chunk, including whether a compressed journaling fragmentation process or a user initiated process emitted the data chunk, a remaining hole size, a size of the remaining data chunk, a ratio between two split pieces of the data chunk, a total size of holes in a storage system, an average compression ratio in the storage system, and a compression ratio in the storage block containing compressed data, for determining whether the data chunk should be split, wherein:
when the data chunk should be split, any possible remaining portion of the data chunk is compressed to a next chronological storage block, and
when the data chunk should not be split, all of the data chunk is moved to the next chronological storage block while any remaining space in the storage block is left as unused.

2. The method of claim 1 , further including, during one of an update operation and a garbage collection operation on the storage block, performing one of:
preventing a splitting of the data chunks thereby leaving holes in the storage block, and
splitting the data chunks and placing the split data chunks into chronological storage blocks.

3. The method of claim 1 , further including preventing a first unused portion of the storage block from being used for write operations until the entire portion of the storage block is freed.

4. The method of claim 1 , further including configuring a second unused portion of the storage block to be used for write operations.

5. The method of claim 1 , further including, performing one of:
writing a portion of the data chunk, and
updating the remaining portion of the data chunk.

6. The method of claim 1 , further including, if the data chunk is unable to be completely compressed into the storage block, performing one of:
writing a portion of the data chunk, and
updating the remaining portion of the data chunk.

7. The method of claim 1 , further including allocating the next chronological storage block.

Assignees

• IBM

Inventors

• Amit Jonathan
• Marenkov Sergey
• Sasson Ben
• Shalev Ori

Classifications

• G06F16/1744 Primary

  using compression, e.g. sparse files · CPC title

• G06F16/1815

  Journaling file systems · CPC title

• G06F16/22

  Indexing; Data structures therefor; Storage structures · CPC title

• G06F17/30153 Primary

  Physics · mapped topic

• G06F17/30312

  Physics · mapped topic