[lustre-devel] RFC: Spill device for Lustre OSD

[Jinshan Xiong]

On Tue, Nov 4, 2025 at 4:46 PM Andreas Dilger <adilger at dilger.ca> wrote:

> On Nov 4, 2025, at 5:28 PM, Jinshan Xiong <jinshanx at google.com> wrote:
> > On Tue, Nov 4, 2025 at 4:20 PM Andreas Dilger wrote:
> >> I don't see how the layout xattr format is related to filesystem-level
> >> scanning?  This is "just" the xattr format *STORED ON THE OST OBJECT*,
> >> and it could be handled by direct device-level (OST) scan tools as well.
> >
> > I see. I initially thought you proposed the spill device would function
> as a mirror and that the existing software stack would handle it.
> >
> > Are we going to store the spilled object's status in the layout? It
> sounds difficult because you will have to initiate a layout change from the
> OSTs. I think we still need to store such an xattr locally in the OST
> object.
>
> I'm exactly proposing that the layout xattr be stored on the OST object.
> It would have the same format as the existing Lustre layout stored on
> the MDT, but *maybe* a different name (e.g. "trusted.spl", "trusted.osd",
> or whatever).
>
> I'm not sure if using "trusted.lov" will confuse filesystem scanning
> tools or help them...  It probably makes sense to have a different
> xattr name, just in case we want MDT spilling (e.g. NVRAM->QLC) in
> the future.
>

I would prefer a different name, even though the client likely won't see it
directly. Using the same name for different things will only lead to
confusion. I would go for "trusted.spl".


>
> >> This would allow mirrors (FLR), multiple spill devices (multiple FLR
> >> mirrors), concatenated devices (PFL), etc.  It just avoids adding new
> >> formats that need to be parsed, printed, etc.
>
> Cheers, Andreas
>
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-devel-lustre.org/attachments/20251104/58075289/attachment.htm>