Just before spring break, I gave a talk at RITlug on the best filesystem in the history of filesystems: ZFS.

While that might be an opinion, it’s certainly cool, and worth talking about! There’s a lot ZFS can do, and a reason it’s sweeping server environments.

I covered the history of ZFS, the features it provides, and some example commands to get started, and plenty of resources to check out! You can check it out on RITlug’s website.

TLDR

ZFS is a copy-on-write (COW) filesystem: every time it writes data, it writes it to a new place, rather than overwriting. The new location for the updated data is written, and all is well. While that sounds a bit wasteful, it means old data can be saved without any cost: just don’t throw away the old location when you write! ZFS tracks these old copies, called snapshots, and lets you jump back to them at a later date, or access any data they hold.

It’s great to take a snapshot (quickly!) before a system upgrade, and roll back if anything breaks.

ZFS has many more “modern” features: supporting huge amounts of data, verifying its integrity, compressing it before writing to disk, deduplication, and many others!

All talk

I enjoy giving talks. Not only can I (hopefully) help break down a complicated subject into bite-sized pieces, but presentation prep teaches me a lot about a subject! It’s also a chance to improve my public speaking and social skills. 🙂

Leave a comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.