- Jul 2023
-
linuxopsys.com linuxopsys.com
-
Btrfs
appears to have an emphasis in security and data-integrity. It its safer when gradual changes in your system are performed. Instead,ext4
appears to lean more to reliability and speed. Backups and deduplication are harder inext4
Also,btrfs
has the ability to create links for duplicate files automatically, liberating disk space.
Tags
Annotators
URL
-
- Oct 2020
-
www.youtube.com www.youtube.com
-
Linux Memory Management at Scale
"we had to build a complete and compliant operating system in order to perform resource control reliably"
epic real-talk. the only people on the planet who seemed to have tamed linux for workloads. controlling memory. taming io. being on the bleeding edge, it turns out, is almost entirely about forward-progress. what can we reclaim?
- oomd for memory protection
- fbtax2
- psi monitoring for io regulation
- cgroups v2
https://facebookmicrosites.github.io/cgroup2/docs/fbtax-results.html
Tags
Annotators
URL
-
- Dec 2019
-
github.com github.com
-
In BTRFS mode, snapshots are taken using the in-built features of the BTRFS filesystem. BTRFS snapshots are supported only on BTRFS systems having an Ubuntu-type subvolume layout (with @ and @home subvolumes).
Tags
Annotators
URL
-
- Dec 2018
-
btrfs.wiki.kernel.org btrfs.wiki.kernel.org
Tags
Annotators
URL
-
- Jun 2017
-
www.usenix.org www.usenix.org
-
zfsembedsdatadirectlyinalogentry.Forlargeval-ues,itwritesdatatodiskredirect-on-write,andstoresapointerinthelog[21].Thisgiveszfsfastdurabilityforsmallwritesbyflushingthelog,avoidstheoverheadofwritinglargevaluestwice,andretainstherecoveryse-manticsofdatajournaling.Ontheotherhand,btrfs[26]usesindirectionforallwrites,regardlessofsize.Itwritesdatatonewly-allocatedblocks,andrecordsthosewriteswithpointersinitsjournal
-
- Nov 2014
-
wiki.ubuntu.com wiki.ubuntu.com
-
several filesystems show performance improvements including XFS and Btrfs
Now using btrfs on my notebook. Works fine!
One of the reasons for using it (from btrfs.wiki.kernel.org):
"SSD (Flash storage) awareness (TRIM/Discard for reporting free blocks for reuse) and optimizations (e.g. avoiding unnecessary seek optimizations, sending writes in clusters, even if they are from unrelated files. This results in larger write operations and faster write throughput)"
Since btrfs once ate my data (pre 1.0) I did several crash simulations to boost my confidence beforehand.
Tags
Annotators
URL
-