From: Miklos Szeredi <miklos@szeredi.hu>
To: Yurii Zubrytskyi <zyy@google.com>
Cc: Eugene Zemtsov <ezemtsov@google.com>,
Amir Goldstein <amir73il@gmail.com>,
linux-fsdevel@vger.kernel.org
Subject: Re: Initial patches for Incremental FS
Date: Thu, 30 May 2019 11:22:54 +0200 [thread overview]
Message-ID: <CAJfpeguys2P9q5EpE3GzKHcOS9GVLO9Fj9HB3JBLw36eax+NkQ@mail.gmail.com> (raw)
In-Reply-To: <CAJeUaNBn0gA6eApgOu=n2uoy+6PbOR_xjTdVvc+StvOKGA-i=Q@mail.gmail.com>
On Wed, May 29, 2019 at 11:06 PM Yurii Zubrytskyi <zyy@google.com> wrote:
> Yes, and this was _exactly_ our first plan, and it mitigates the read
> performance
> issue. The reasons why we didn't move forward with it are that we figured out
> all other requirements, and fixing each of those needs another change in
> FUSE, up to the level when FUSE interface becomes 50% dedicated to
> our specific goal:
> 1. MAP message would have to support data compression (with different
> algorithms), hash verification (same thing) with hash streaming (because
> even the Merkle tree for a 5GB file is huge, and can't be preloaded
> at once)
With the proposed FUSE solution the following sequences would occur:
kernel: if index for given block is missing, send MAP message
userspace: if data/hash is missing for given block then download data/hash
userspace: send MAP reply
kernel: decompress data and verify hash based on index
The kernel would not be involved in either streaming data or hash, it
would only work with data/hash that has already been downloaded.
Right?
Or is your implementation doing streamed decompress/hash or partial blocks?
> 1.1. Mapping memory usage can get out of hands pretty quickly: it has to
> be at least (offset + size + compression type + hash location + hash size +
> hash kind) per each block. I'm not even thinking about multiple storage files
> here. For that 5GB file (that's a debug APK for some Android game we're
> targeting) we have 1.3M blocks, so ~16 bytes *1.3M = 20M of index only,
> without actual overhead for the lookup table.
> If the kernel code owns and manages its own on-disk data store and the
> format, this index can be loaded and discarded on demand there.
Why does the kernel have to know the on-disk format to be able to load
and discard parts of the index on-demand? It only needs to know which
blocks were accessed recently and which not so recently.
> > There's also work currently ongoing in optimizing the overhead of
> > userspace roundtrip. The most promising thing appears to be matching
> > up the CPU for the userspace server with that of the task doing the
> > request. This can apparently result in 60-500% speed improvement.
>
> That sounds almost too good to be true, and will be really cool.
> Do you have any patches or git remote available in any compilable state to
> try the optimization out? Android has quite complicated hardware config
> and I want to see how this works, especially with our model where
> several processes may send requests into the same filesystem FD.
Currently it's only a bunch of hacks, no proper interfaces yet.
I'll let you know once there's something useful for testing with a
real filesystem.
BTW, which interface does your fuse filesystem use? Libfuse? Raw device?
Thanks,
Miklos
next prev parent reply other threads:[~2019-05-30 9:23 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-02 4:03 Initial patches for Incremental FS ezemtsov
2019-05-02 4:03 ` [PATCH 1/6] incfs: Add first files of incrementalfs ezemtsov
2019-05-02 19:06 ` Miklos Szeredi
2019-05-02 20:41 ` Randy Dunlap
2019-05-07 15:57 ` Jann Horn
2019-05-07 17:13 ` Greg KH
2019-05-07 17:18 ` Greg KH
2019-05-02 4:03 ` [PATCH 2/6] incfs: Backing file format ezemtsov
2019-05-02 4:03 ` [PATCH 3/6] incfs: Management of in-memory FS data structures ezemtsov
2019-05-02 4:03 ` [PATCH 4/6] incfs: Integration with VFS layer ezemtsov
2019-05-02 4:03 ` [PATCH 6/6] incfs: Integration tests for incremental-fs ezemtsov
2019-05-02 11:19 ` Initial patches for Incremental FS Amir Goldstein
2019-05-02 13:10 ` Theodore Ts'o
2019-05-02 13:26 ` Al Viro
2019-05-03 4:23 ` Eugene Zemtsov
2019-05-03 5:19 ` Amir Goldstein
2019-05-08 20:09 ` Eugene Zemtsov
2019-05-09 8:15 ` Amir Goldstein
[not found] ` <CAK8JDrEQnXTcCtAPkb+S4r4hORiKh_yX=0A0A=LYSVKUo_n4OA@mail.gmail.com>
2019-05-21 1:32 ` Yurii Zubrytskyi
2019-05-22 8:32 ` Miklos Szeredi
2019-05-22 17:25 ` Yurii Zubrytskyi
2019-05-23 4:25 ` Miklos Szeredi
2019-05-29 21:06 ` Yurii Zubrytskyi
2019-05-30 9:22 ` Miklos Szeredi [this message]
2019-05-30 22:45 ` Yurii Zubrytskyi
2019-05-31 9:02 ` Miklos Szeredi
2019-05-22 10:54 ` Amir Goldstein
2019-05-03 7:23 ` Richard Weinberger
2019-05-03 10:22 ` Miklos Szeredi
2019-05-02 13:46 ` Amir Goldstein
2019-05-02 18:16 ` Richard Weinberger
2019-05-02 18:33 ` Richard Weinberger
2019-05-02 13:47 ` J. R. Okajima
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAJfpeguys2P9q5EpE3GzKHcOS9GVLO9Fj9HB3JBLw36eax+NkQ@mail.gmail.com \
--to=miklos@szeredi.hu \
--cc=amir73il@gmail.com \
--cc=ezemtsov@google.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=zyy@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).