#filesystems

[x11-official]

The default way of sorting files should really be "the newest of modification/creation" instead of just modification time.

At work I'll download an image and then take a couple minutes to find it in the huge "Pictures" folder because it was "last modified" some time in 2023.

Of course some filesystems don't support creation time. But the ones most commonly used on Linux today (ext4/btrfs/XFS/ZFS) do, as does Windows's NTFS and Mac's APFS.

This also gets back into my idea that the filesystem could treat "files meant to be edited" and "files meant to be atomic" differently.

[radicalityincident]

Something that I think is really fundamental about what appeals to people about things like Linux and Firefox and other open source and customizable software is that a lot of us remember the days when we told computers what to do. These days it really feels like we have decided to let computers tell us what to do (and by extension nazi fuck tech billionaires).

Don't let the computer tell you what to do. You tell the computer what to do! I always remember when I was newer to Linux I wanted advice on where to store files that I wanted multiple users to have easy read write access to, so I asked and people were generally just like "I dunno, I guess /opt or /srv. It doesn't really matter its your filesystem" and that really spoke to me. Like, its your filesystem, its your computer. Create a directory on your fs root called /stuff and just dump everything you have in there if you want. Create different partitions for files that are different colors if you want. Store all your nudes in /proc no one gives a shit.

[brightgreendandelions]

alright i have a ✨Diabolical Plan ✨

instead of just installing arch normally, i'm gonna boot into the install iso, format the empty partition i created from win10 as ext4, use fbcat(1) or something to make a screenshot, save it to that partition, and then try to mount that from WSL to post the screenshot on tumblr.

[hpcjourney]

GPFS - Investigating - RPC Wait

I am looking into the 'RPC wait' for node getData with a named node. I am trying to understand how this error message works, and the internals of what it means.

Update 10/13/25:

Well it's kinda opaque how this all works. It also seems like there are a couple possible reasons for why this could be happening. It was not clear to me if there is a way to tell, in the abstract, which one without looking at node logs.

It seems like the most likely reason is RDMA issues. Likely due to network issues where the communications are not getting through as they should causing 'RPC wait' issues.

[innovatexblog]

Software & OS Impacts: Memory Management, Page Size, & Filesystems

Explore how OS design affects memory, page size, and filesystems—key factors shaping performance, stability, and hardware compatibility.

💾 Why Your Hardware Is Only Half the Story... Speed doesn’t just come from faster RAM or SSDs — your operating system plays a critical role in how memory and storage are actually used.

🧠 In this blog, we break down: • How OS memory management impacts speed and efficiency • Why page size can make or break performance • The secret power of filesystems in storage handling

If you’ve ever wondered why high-end hardware still lags — this post has the answers.

🔗 Read the full article now and uncover how software architecture drives performance 

[dvdmerwe]

“ZFS and Better File System (Btrfs) are two popular modern file systems. Both are designed with advanced storage features and offer snapshot capabilities, data integrity protection, and highly efficient storage management. However, ZFS has been the more reliable option for both personal and enterprise-level usage.”

I found this interesting as I did try a switch to Btrfs myself a while ago, but seem to remember I had speed issues as well as an inflated storage requirement. I liked the concept though. It is something I want to revisit in the future again, but I’ll probably give ZFS a more serious consideration then.

Right now I’m back on EXT4 with daily mirrored backups to two separate locations, and a selected cloud backup of critical data.

See https://www.xda-developers.com/how-zfs-is-superior-to-btrfs

[hpcjourney]

GPFS - Notes - Install Internals

You can actually try IBM Storage Scale for free–obviously there are limits, but you can certainly get a sense of how it all works. To that end I was pulling apart the installer the other day and noticed that I couldn't find mmdiag.

I was confused by this–until I unpacked the .deb file and looked at the post install script. As it turns out IBM makes liberal use of symlinks and mmdiag is just a symlink to tsdiag. I wanted to document that here–so I don't forget and in case anyone else finds themselves needlessly puzzled. I also posted a list of the symlinks I found in the GPFS Base .deb file here.

I am currently trying to understand the various diagnostic tools built into GPFS, and I wanted to look at how things work internally.

[hpcjourney]

Research Topic - DAOS

You can download the source here:

GitHub - daos-stack/daos: DAOS Storage Stack (client libraries, storage engine, control plane)

DAOS Storage Stack (client libraries, storage engine, control plane) - daos-stack/daos

Just recording this here to research later.