[PATCH 0/3] blk-throttle: async write throttling

Andrea Righi arighi at develer.com
Mon Feb 28 02:15:02 PST 2011

Currently the blkio.throttle controller only support synchronous IO requests.
This means that we always look at the current task to identify the "owner" of
each IO request.

However dirty pages in the page cache can be wrote to disk asynchronously by
the per-bdi flusher kernel threads or by any other thread in the system,
according to the writeback policy.

For this reason the real writes to the underlying block devices may
occur in a different IO context respect to the task that originally
generated the dirty pages involved in the IO operation. This makes the
tracking and throttling of writeback IO more complicate respect to the
synchronous IO from the blkio controller's perspective.

Proposed solution
In the previous patch set http://lwn.net/Articles/429292/ I proposed to resolve
the problem of the buffered writes limitation by tracking the ownership of all
the dirty pages in the system.

This would allow to always identify the owner of each IO operation at the block
layer and apply the appropriate throttling policy implemented by the
blkio.throttle controller.

This solution makes the blkio.throttle controller to work as expected also for
writeback IO, but it does not resolve the problem of faster cgroups getting
blocked by slower cgroups (that would expose a potential way to create DoS in
the system).

In fact, at the moment critical IO requests (that have dependency with other IO
requests made by other cgroups) and non-critical requests are mixed together at
the filesystem layer in a way that throttling a single write request may stop
also other requests in the system, and at the block layer it's not possible to
retrieve such informations to make the right decision.

A simple solution to this problem could be to just limit the rate of async
writes at the time a task is generating dirty pages in the page cache. The
big advantage of this approach is that it does not need the overhead of
tracking the ownership of the dirty pages, because in this way from the blkio
controller perspective all the IO operations will happen from the process
context: writes in memory and synchronous reads from the block device.

The drawback of this approach is that the blkio.throttle controller becomes a
little bit leaky, because with this solution the controller is still affected
by the IO spikes during the writeback of dirty pages executed by the kernel

Probably an even better approach would be to introduce the tracking of the
dirty page ownership to properly account the cost of each IO operation at the
block layer and apply the throttling of async writes in memory only when IO
limits are exceeded.

To summarize, we can identify three possible solutions to properly throttle the
buffered writes:

1) account & throttle everything at block IO layer (bad for "priority
   inversion" problems, needs page tracking for blkio)

2) account at block IO layer and throttle in memory (needs page tracking for

3) account & throttle in memory (affected by IO spikes, depending on
   dirty_ratio / dirty_background_ratio settings)

For now we start with the solution 3) that seems to be the simplest way to

- create a cgroup with 4MiB/s write limit:
  # mount -t cgroup -o blkio none /mnt/cgroup
  # mkdir /mnt/cgroup/foo
  # echo 8:0 $((4 * 1024 * 1024)) > /mnt/cgroup/foo/blkio.throttle.write_bps_device

  NOTE: async io is still limited per-device, as well as sync io

- move a task into the cgroup and run a dd to generate some writeback IO


  - 2.6.38-rc6 vanilla:

  $ cat /proc/$$/cgroup
  $ dd if=/dev/zero of=zero bs=1M count=128 &
  $ dstat -df
   read  writ
   0     0
   0     0
   0    22M  <--- writeback starts here and is not limited at all
   0    43M
   0    45M
   0    18M

  - 2.6.38-rc6 + async write throttling:

  $ cat /proc/$$/cgroup
  $ dd if=/dev/zero of=zero bs=1M count=128 &
  $ dstat -df
   read  writ
   0     0
   0     0
   0     0
   0     0
   0    22M  <--- we have some IO spikes but the overall writeback IO
   0     0        is controlled according to the blkio write limit
   0     0
   0     0
   0     0
   0    29M
   0     0
   0     0
   0     0
   0     0
   0    26M
   0     0
   0     0
   0     0
   0     0
   0    30M
   0     0
   0     0
   0     0
   0     0
   0    20M

 - Consider to add the following new files in the blkio controller to allow the
   user to explicitly limit async writes as well as sync writes:


Any feedback is welcome.

[PATCH 1/3] block: introduce REQ_DIRECT to track direct io bio
[PATCH 2/3] blkio-throttle: infrastructure to throttle async io
[PATCH 3/3] blkio-throttle: async write io instrumentation

 block/blk-throttle.c      |  106 ++++++++++++++++++++++++++++++---------------
 fs/direct-io.c            |    1 +
 include/linux/blk_types.h |    2 +
 include/linux/blkdev.h    |    6 +++
 mm/page-writeback.c       |   17 +++++++
 5 files changed, 97 insertions(+), 35 deletions(-)

More information about the Containers mailing list