Linux-mm Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Lance Yang <lance.yang@linux.dev>
To: sj@kernel.org, mst@redhat.com, david@kernel.org
Cc: damon@lists.linux.dev, linux-mm@kvack.org,
	virtualization@lists.linux.dev, ryncsn@gmail.com,
	kunwu.chan@gmail.com, lianux.mm@gmail.com, baohua@kernel.org,
	xueyuan.chen21@gmail.com, Lance Yang <lance.yang@linux.dev>
Subject: [FYI] DAMON and virtio-balloon in DeepSeek's DSec paper
Date: Fri, 25 Sep 2026 13:44:08 +0800	[thread overview]
Message-ID: <20260925054408.10431-1-lance.yang@linux.dev> (raw)

Hi all,

I was reading DeepSeek's new DSec paper[1] and found a nice use of DAMON
and virtio-balloon:

With this kind of workload, an agent may read a file once and never touch
it again, while those pages remain in the guest page cache. Without memory
pressure in the guest, they can stay cached even though the host would
like that memory back ...

The trick is DAMON + virtio-balloon free-page reporting :) DAMON reclaims
cold file pages from the guest page cache; buddy gets a chance to coalesce
them into reportable blocks, and virtio-balloon passes those blocks to
Firecracker. Firecracker can then drop the host backing with MADV_DONTNEED.

And it worked: time-integrated host memory use fell by 21.2%, while peak
usage stayed about the same, with no significant CPU overhead. Pretty cool
to see these pieces show up in another real, large-scale system :)

Kernel work still changes the world. Cheers to that!

Anyhow, I still don't know DAMON and virtio-balloon as well as I should
(mostly I just know the people who built them) :P

[1] https://arxiv.org/html/2609.22978

Cheers, Lance


             reply	other threads:[~2026-09-25  5:44 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-25  5:44 Lance Yang [this message]
2026-09-25  5:57 ` [FYI] DAMON and virtio-balloon in DeepSeek's DSec paper Lian Wang
2026-09-25  6:46   ` KunWu Chan
2026-09-25  7:04 ` David Hildenbrand (Arm)
2026-09-25  7:28   ` Lian Wang
2026-09-25  8:15   ` Lance Yang
2026-09-25 10:03     ` David Hildenbrand (Arm)
2026-09-25 10:16     ` Gao Xiang
2026-09-25 10:27       ` Gao Xiang
2026-09-25 10:13 ` SJ Park
2026-09-29 12:32 ` Jialiang Huang
2026-09-29 12:41   ` Gao Xiang
2026-09-29 12:50     ` Jialiang Huang
2026-09-30  3:36     ` Muchun Song
2026-09-30  7:24       ` Gao Xiang
2026-09-30  9:37         ` Muchun Song
2026-09-30 10:33           ` Gao Xiang
2026-09-29 16:56   ` SJ Park
2026-09-30  3:07     ` Lian Wang
2026-09-30  8:08       ` SJ Park
2026-09-29 18:14   ` Pratyush Mallick

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=20260925054408.10431-1-lance.yang@linux.dev \
    --to=lance.yang@linux.dev \
    --cc=baohua@kernel.org \
    --cc=damon@lists.linux.dev \
    --cc=david@kernel.org \
    --cc=kunwu.chan@gmail.com \
    --cc=lianux.mm@gmail.com \
    --cc=linux-mm@kvack.org \
    --cc=mst@redhat.com \
    --cc=ryncsn@gmail.com \
    --cc=sj@kernel.org \
    --cc=virtualization@lists.linux.dev \
    --cc=xueyuan.chen21@gmail.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