Timeline for How to append a chunk of fixed size data to a file and make sure this chunk doesn't get fragmented on disk?
Current License: CC BY-SA 4.0
18 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 16, 2020 at 19:58 | vote | accept | OneAndOnly | ||
| Jan 13, 2020 at 23:37 | answer | added | Anton Astafiev | timeline score: 1 | |
| Dec 31, 2018 at 14:00 | vote | accept | OneAndOnly | ||
| Jan 16, 2020 at 19:58 | |||||
| Dec 21, 2018 at 21:23 | comment | added | amon | Disk fragmentation is not user-visible, but abstracted away by the file system. Modern file systems defrag automatically in the background. Also, disk fragmentation primarily affects sequential access time on hard disks. That's not necessarily relevant for a database, especially not when using SSDs or RAID arrays. Speaking of which, SSDs often split up data internally for wear leveling. So the logical block device that the operating system sees usually has little relationship to the physical storage of the data. | |
| Dec 21, 2018 at 4:35 | answer | added | Kain0_0 | timeline score: 0 | |
| Dec 20, 2018 at 16:26 | comment | added | Erik Eidt | One approach is to create a large file in advance, then defragement it (using one of the standard approaches), then write to it (e.g. starting at offset 0, as if it was blank). docs.microsoft.com/en-us/windows/desktop/fileio/… | |
| Dec 20, 2018 at 14:52 | history | edited | OneAndOnly | CC BY-SA 4.0 |
added 192 characters in body
|
| Dec 20, 2018 at 14:46 | history | edited | OneAndOnly | CC BY-SA 4.0 |
added 102 characters in body
|
| Dec 20, 2018 at 14:45 | answer | added | Caleth | timeline score: 2 | |
| Dec 20, 2018 at 14:44 | comment | added | OneAndOnly | @TheCatWhisperer but based on papers that I'm reading, all of the relational databases use these pages and dont get fragmented, and im not asking about a specific database or language, just any tips on how its done. paper : sciencedirect.com/science/article/pii/S1742287615000584 | |
| Dec 20, 2018 at 14:43 | comment | added | TheCatWhisperer | @OneAndOnly, this question is off topic here because it is asking about a specific implementation rather than general engineering discipline. With that said, it's a good question, and now I am very curious to the answer! | |
| Dec 20, 2018 at 14:38 | history | edited | OneAndOnly | CC BY-SA 4.0 |
added 461 characters in body
|
| Dec 20, 2018 at 14:36 | comment | added | OneAndOnly | @Neil I know, my point is lets say O.S stores files based on 4KB chunks, and may fragment some files when they exceed it, and the DBMS uses 16KB pages, my question is how do they implement that DBMS so that 16KB pages which get added to table files, dont get fragmented? when i append a 16KB data to a file, is it by default reserved for it and will never get fragmented? | |
| Dec 20, 2018 at 14:22 | comment | added | Neil | Generally files are saved in chunks of 4096KB (OS-dependent however). If the file doesn't exceed this amount, it will never be stored on multiple chunks. More reading here. | |
| Dec 20, 2018 at 14:20 | review | Close votes | |||
| Jan 5, 2019 at 3:05 | |||||
| Dec 20, 2018 at 14:08 | history | edited | OneAndOnly | CC BY-SA 4.0 |
added 244 characters in body; edited title
|
| Dec 20, 2018 at 14:00 | review | First posts | |||
| Dec 21, 2018 at 3:44 | |||||
| Dec 20, 2018 at 14:00 | history | asked | OneAndOnly | CC BY-SA 4.0 |