Ubuntu 19.10 To Boot Faster Thanks To LZ4 Compression

[skeevy420]

Sorry, but your tl;dr is moot. You compare it to xz which is not what we're talking about.
Anyhow, i did look at your link and you do get very impressive results!
I'm not adventurous enough to try out kernel patches so i just take your word for it

To be clear though. Booting my machine with nvme is already insanely fast in mere seconds (cold boot to login). I'm very sure the initramfs part takes about half a second.. So ehh :P
Yeah well, it's fast. Compression could just make it a tiny bit faster, but still nothing i'd actually notice i think.

On my PC with spinning drives, ZSTD does make a difference over XZ, but barely noticeable but enough to mention it because it could matter for really low end devices.

ZSTD compared to LZ4 on boot, can't tell a difference between the two until I look at raw numbers from benchmarks.

It's how well ZSTD compresses at the really high --fast modes that is really peaking my interests. My ramdisks would love that.

[skeevy420]

markg85 I accidentally left out a key phrase of my first post. It was supposed to be "ZSTD has ~half the compression speed and same decompression speed as LZ4 when tuned with --fast=2 to --fast=4 with compression ratios that equal xz." Sorry about that.

Michael
Unapproved post.

[bug77]

As media gets faster, the load time difference between GZIP, LZ4 and LZO diminishes and the decompression time becomes the dominant speed factor with LZ4 the clear winner.

Weird, I thought when the decompression time holds you back, the best solution would be not to compress in the first place?

[Chaython]

So this is a non-transparent compression? How does the system know to decompress the kernel?
How does this compare to LZX in NTFS?

[ext73]

it can be a long time - my kernel builds have been compressing for 5 years, additionally lz4 for zswap

[FireBurn]

I still use xz for my kernels, might try and test lz4 to see what the difference is

[tuxd3v]

LZ4, is indeed blazing fast,
And in ARMv8 at least its like crazy compared even with ZSTD, I already saw some comparisons and its Amazing..
ZSTD, has the advantage of using some sort of a garbage collector?Maybe I interpreted in wrong, but it seems it has some sort of re-usability advantage..
But anyway, at least on ARMv8 LZ4 is faster, and it also can use a dictionary if needed..

[AndyChow]

On my PC with spinning drives, ZSTD does make a difference over XZ, but barely noticeable but enough to mention it because it could matter for really low end devices.

ZSTD compared to LZ4 on boot, can't tell a difference between the two until I look at raw numbers from benchmarks.

It's how well ZSTD compresses at the really high --fast modes that is really peaking my interests. My ramdisks would love that.

I'd like to know what your setup is, since last time I checked ZSTD isn't supported on boot.

[Danny3]

This is amazing news!
The kind of improvement that I was expecting on a new version of Ubuntu.
This will be very good also for those cases when something crashes the computer and it needs to boot as fast as possible and resume the work.

[skeevy420]

I'd like to know what your setup is, since last time I checked ZSTD isn't supported on boot.

You're right, it isn't, which is why Ieft a link to the patches I made for kernels 4.19 to 5.3 with instructions on how to use it them when building a kernel from Tk-Glitch

An odd Google search about BTRFS ZSTD compression led me to the LKML which led me to the github repo of the guy trying to get ZSTD support in the kernel and then, I think this is the correct one from my history, ran:

git format-patch -3 40654c08a9bfd6455411f2b21bdec74b9df59c09 --stdout > zstd.patch

after I cloned the repo.