GPN16:An Introduction to ZFS: Unterschied zwischen den Versionen

aus dem Wiki des Entropia e.V., CCC Karlsruhe
KKeine Bearbeitungszusammenfassung
K (Video added)
 
(3 dazwischenliegende Versionen von 3 Benutzern werden nicht angezeigt)
Zeile 1: Zeile 1:
{{#css:GPN16:Stylesheet}}
{{#ev:mediacccde|https://media.ccc.de/browse/conferences/gpn/gpn16/gpn16-7633-an_introduction_to_zfs.html||right}}
 
Ein Vortrag von problame auf der [[GPN16]].
Ein Vortrag von problame auf der [[GPN16]].


Zeile 11: Zeile 10:
attempting to solve the problems mentioned above.
attempting to solve the problems mentioned above.


Beginning with an analysis of the deficiencies of current file systems we walk through different technologies that could be used to for mitigation.
Beginning with an analysis of the deficiencies of current file systems & storage stacks, we continue with key design techniques of ZFS.
A short history of the implementation of ZFS is followed by several demos of features that are unique to ZFS / similar systems.
 
Afterwards, the individual modules that form ZFS are presented in more detail (interface, functionality, ...).
This chapter is summarized by following the path a write(3) call takes from userland to actual bits on disk.
 
Finally, a demo of ZFS pool creation and file system basics is given.
 
We finish with a recap & time for questions.
We finish with a recap & time for questions.


Those interested in further exploring ZFS are encouraged to participate
Those interested in further exploring ZFS are encouraged to participate
in the workshop taking place later the same day.
in the workshop taking place later the same day.


== Links ==
== Links ==


 
* [https://perma.cschwarz.com/talks/introzfs2016 Slides (PDF & LaTeX Sources)]


{{Navigationsleiste GPN16:Vorträge}}
{{Navigationsleiste GPN16:Vorträge}}

Aktuelle Version vom 25. Juni 2016, 15:05 Uhr

Ein Vortrag von problame auf der GPN16.

Most file systems used today lack support for many advanced features, e.g. data integrity & low-cost snapshots. Not too long ago, power-outages in the middle of a write could have catastrophic consequences. Volume Managers that do not know about their volumes' file systems are inflexible. Hardware-RAIDs are problems rather than solutions.

This talk gives an overview of ZFS, a (not so) new approach to file systems & volume management, attempting to solve the problems mentioned above.

Beginning with an analysis of the deficiencies of current file systems & storage stacks, we continue with key design techniques of ZFS.

Afterwards, the individual modules that form ZFS are presented in more detail (interface, functionality, ...). This chapter is summarized by following the path a write(3) call takes from userland to actual bits on disk.

Finally, a demo of ZFS pool creation and file system basics is given.

We finish with a recap & time for questions.

Those interested in further exploring ZFS are encouraged to participate in the workshop taking place later the same day.

Links