Data access performance using decompression maps

• US 10,437,826 B2
• Filed: 05/24/2017
• Issued: 10/08/2019
• Est. Priority Date: 09/19/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Lyle, Robert W.
• Primary Examiner(s): Morrison, Jay A
• Assistant Examiner(s): Hoang, Ken
• Application Number:
  US15/604,210
• Publication Number:
  US 20170255678A1
• Time in Patent Office: 867 Days
• Field of Search: 707693, 707748, 707E17005, 707E17018, 707E17084, 707640, 707661, 707674, 707696, 707792, 707999001, 707999002, 707999003, 707999004, 707999005, 707999006, 707999009, 7079991, 707999101, 707999102, 707999103, 707999104, 707999107
• CPC Class Codes: G06F 16/1744 using compression, e.g. sparse files
• CPC Class Codes: G06F 16/2365 Ensuring data consistency and integrity
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 16/24561 Intermediate data storage techniques for performance improvement

Abstract

Techniques for decompressing data in a database system. A query is received, which pertains to a subset of data within a compressed set of data. One or more decompression strategies are evaluated using a cost model. The cost model includes an estimated filter factor. A low cost decompression strategy is selected based on the results of the evaluation of the one or more decompression strategies. One or more bytes representing the requested subset of data are located within the compressed set of data. Only a portion of the compressed data that corresponds to the subset of data is decompressed, using the selected decompression strategy, while leaving the remaining set of data in a compressed state.

Claims

1. A computer program product for decompressing data in a database system, the computer program product comprising a non-transitory computer readable storage medium having program code embodied therewith, the program code being executable by a processor to:

receive, by the processor, a query pertaining to a subset of data within a compressed set of data;

evaluate, by the processor, one or more decompression strategies by calculating a number of cycles needed for decompressing only a subset of columns within a row and comparing the calculated number of cycles with a number of cycles needed for decompressing all columns within the row, and applying an estimated filter factor to the calculated numbers;

select, by the processor, a low cost decompression strategy based on the results of the evaluation of the one or more decompression strategies;

locate, by the processor and by using a decompression map for each row, one or more bytes representing the requested subset of data within the compressed set of data, wherein the decompression map is stored as an array and describes the bytes representing a variable length column instance within the row, and wherein the subset of data includes one of;

a subset of columns within a row of a database table, a subset of bytes within a row of the database table, and a subset of rows within a page of the database table; and

decompress, by the processor, only a portion of the compressed data that corresponds to the subset of data, using the selected decompression strategy, while leaving the remaining set of data in a compressed state.

2. The computer program product of claim 1, wherein the decompression strategy includes two or more decompression phases, and wherein a first decompression phase includes decompression of the columns needed for predicate evaluation and a second decompression phase includes decompression of all remaining columns needed for a select list.

3. The computer program product of claim 1, wherein locating one or more bytes representing the requested subset of data within the compressed set of data includes:

determining, by the processor, a unit representing a certain number of bytes of compressed data; and

determining, by the processor, what units need to be decompressed to get to the one or more bytes representing the requested subset of data within the compressed set of data.

4. The computer program product of claim 3, wherein every variable length column is preceded by a number describing the length of the variable length column.

5. The computer program product of claim 1, wherein decompressing starts at the beginning of a row.

6. The computer program product of claim 1, wherein decompressing starts at the end of a row.