mbox series

[v10,00/16] Introduce Data Access MONitor (DAMON)

Message ID 20200505110815.10532-1-sjpark@amazon.com (mailing list archive)
Headers show
Series Introduce Data Access MONitor (DAMON) | expand

Message

SeongJae Park May 5, 2020, 11:07 a.m. UTC
From: SeongJae Park <sjpark@amazon.de>

Introduction
============

Memory management decisions can be improved if finer data access information is
available.  However, because such finer information usually comes with higher
overhead, most systems including Linux forgives the potential benefit and rely
on only coarse information or some light-weight heuristics.  The pseudo-LRU and
the aggressive THP promotions are such examples.

A number of data access pattern awared memory management optimizations (refer
to 'Appendix A' for more details) consistently say the potential benefit is not
small.  However, none of those has successfully merged to the mainline Linux
kernel mainly due to the absence of a scalable and efficient data access
monitoring mechanism.  Refer to 'Appendix B' to see the limitations of existing
memory monitoring mechanisms.

DAMON is a data access monitoring subsystem for the problem.  It is 1) accurate
enough to be used for the DRAM level memory management (a straightforward
DAMON-based optimization achieved up to 2.55x speedup), 2) light-weight enough
to be applied online (compared to a straightforward access monitoring scheme,
DAMON is up to 94,242.42x lighter) and 3) keeps predefined upper-bound overhead
regardless of the size of target workloads (thus scalable).  Refer to 'Appendix
C' if you interested in how it is possible, and 'Appendix F' to know how the
numbers collected.

DAMON has mainly designed for the kernel's memory management mechanisms.
However, because it is implemented as a standalone kernel module and provides
several interfaces, it can be used by a wide range of users including kernel
space programs, user space programs, programmers, and administrators.  DAMON
is now supporting the monitoring only, but it will also provide simple and
convenient data access pattern awared memory managements by itself.  Refer to
'Appendix D' for more detailed expected usages of DAMON.


Boring? Here Are Something Colorful
===================================

For intuitive understanding of DAMON, I made web pages[1-8] showing the
visualized dynamic data access pattern of various realistic workloads in
PARSEC3 and SPLASH-2X bechmark suites.  The figures are generated using the
user space tool in 10th patch of this patchset.

There are pages showing the heatmap format dynamic access pattern of each
workload for heap area[1], mmap()-ed area[2], and stack[3] area.  I splitted
the entire address space to the three area because there are huge unmapped
regions between the areas.

You can also show how the dynamic working set size of each workload is
distributed[4], and how it is chronologically changing[5].

The most important characteristic of DAMON is its promise of the upperbound of
the monitoring overhead.  To show whether DAMON keeps the promise well, I
visualized the number of monitoring operations required for each 5
milliseconds, which is configured to not exceed 1000.  You can show the
distribution of the numbers[6] and how it changes chronologically[7].

[1] https://damonitor.github.io/reports/latest/by_image/heatmap.0.png.html
[2] https://damonitor.github.io/reports/latest/by_image/heatmap.1.png.html
[3] https://damonitor.github.io/reports/latest/by_image/heatmap.2.png.html
[4] https://damonitor.github.io/reports/latest/by_image/wss_sz.png.html
[5] https://damonitor.github.io/reports/latest/by_image/wss_time.png.html
[6] https://damonitor.github.io/reports/latest/by_image/nr_regions_sz.png.html
[7] https://damonitor.github.io/reports/latest/by_image/nr_regions_time.png.html


Future Plans
============

This patchset is only for the first stage of DAMON.  As soon as this patchset
is merged, official patchsets for below future plans will be posted.


Automate Data Access Pattern-aware Memory Management
----------------------------------------------------

Though DAMON provides the monitoring feature, implementing data access pattern
aware memory management schemes could be difficult to beginners.  DAMON will be
able to do most of the work by itself in near future.  Users will be required
to only describe what kind of data access monitoring-based operation schemes
they want.

By applying a very simple scheme for THP promotion/demotion with a latest
version of the patchset (not posted yet), DAMON achieved 18x lower memory space
overhead compared to THP while preserving about 50% of the THP performance
benefit with SPLASH-2X benchmark suite.

An RFC patchset for this plan is already available
(https://lore.kernel.org/linux-mm/20200429124540.32232-1-sjpark@amazon.com/).


Support Various Address Spaces
------------------------------

Currently, DAMON supports virtual memory address spaces using PTE Accessed bits
as its access checking primitive.  However, the core design of DAMON is not
dependent to such implementation details.  In a future, DAMON will decouple
those and support various address spaces including physical memory.  It will
further allow users to configure and even implement the primitives by
themselves for their special usecase.  Monitoring of page cache, NUMA nodes,
specific files, or block devices would be examples of such usecases.

An RFC patchset for this plan is already available
(https://lore.kernel.org/linux-mm/20200409094232.29680-1-sjpark@amazon.com/).


Frequently Asked Questions
==========================

Q: Why a new module, instead of extending perf or other tools?
A: First, DAMON aims to be used by other programs including the kernel.
Therefore, having dependency to specific tools like perf is not desirable.
Second, because it need to be lightweight as much as possible so that it can be
used online, any unnecessary overhead such as kernel - user space context
switching cost should be avoided.  These are the two most biggest reasons why
DAMON is implemented in the kernel space.  The idle page tracking subsystem
would be the kernel module that most seems similar to DAMON.  However, its own
interface is not compatible with DAMON.  Also, the internal implementation of
it has no common part to be reused by DAMON.

Q: Can 'perf mem' or PMUs used instead of DAMON?
A: No.  Roughly speaking, DAMON has two seperate layers.  The low layer is
access check of pages, and the higher layer is its core mechanisms for overhead
controlling.  For the low layer, DAMON is now using the PTE Accessed bits.
Other H/W or S/W features that can be used for the access check of pages, such
as 'perf mem', PMU, or even page idle, could be used instead in the layer.
However, those could not alternate the high layer of DAMON.


Evaluations
===========

We evaluated DAMON's overhead, monitoring quality and usefulness using 25
realistic workloads on my QEMU/KVM based virtual machine.

DAMON is lightweight.  It consumes only -0.03% more system memory and up to 1%
CPU time.  It makes target worloads only 1.02% slower.

DAMON is accurate and useful for memory management optimizations.  An
experimental DAMON-based operation scheme for THP removes 88.43% of THP memory
overheads while preserving 36.36% of THP speedup.  Another experimental
DAMON-based 'proactive reclamation' implementation reduces 86.37% of
residentail sets and 22% of system memory footprint while incurring only 2.03%
runtime overhead in best case (parsec3/freqmine).

NOTE that the experimentail THP optimization and proactive reclamation are not
for production, just only for proof of concepts.

Please refer to 'Appendix E' for detailed evaluation setup and results.


References
==========

Prototypes of DAMON have introduced by an LPC kernel summit track talk[1] and
two academic papers[2,3].  Please refer to those for more detailed information,
especially the evaluations.  The latest version of the patchsets has also
introduced by an LWN artice[4].

[1] SeongJae Park, Tracing Data Access Pattern with Bounded Overhead and
    Best-effort Accuracy. In The Linux Kernel Summit, September 2019.
    https://linuxplumbersconf.org/event/4/contributions/548/
[2] SeongJae Park, Yunjae Lee, Heon Y. Yeom, Profiling Dynamic Data Access
    Patterns with Controlled Overhead and Quality. In 20th ACM/IFIP
    International Middleware Conference Industry, December 2019.
    https://dl.acm.org/doi/10.1145/3366626.3368125
[3] SeongJae Park, Yunjae Lee, Yunhee Kim, Heon Y. Yeom, Profiling Dynamic Data
    Access Patterns with Bounded Overhead and Accuracy. In IEEE International
    Workshop on Foundations and Applications of Self- Systems (FAS 2019), June
    2019.
[4] Jonathan Corbet, Memory-management optimization with DAMON. In Linux Weekly
    News (LWN), Feb 2020. https://lwn.net/Articles/812707/


Baseline and Complete Git Trees
===============================

The patches are based on the v5.6.  You can also clone the complete git
tree:

    $ git clone git://github.com/sjp38/linux -b damon/patches/v10

The web is also available:
https://github.com/sjp38/linux/releases/tag/damon/patches/v10

This patchset contains patches for the stabled main logic of DAMON only.  The
latest DAMON development tree is also available at:
https://github.com/sjp38/linux/tree/damon/master


Sequence Of Patches
===================

The patches are organized in the following sequence.  The first two patches are
preparation of DAMON patchset.  The 1st patch adds typos found in previous
versions of DAMON patchset to 'scripts/spelling.txt' so that the typos can be
caught by 'checkpatch.pl'.  The 2nd patch exports 'lookup_page_ext()' to GPL
modules so that it can be used by DAMON even though it is built as a loadable
module.

Next five patches implement the core of DAMON and it's programming interface.
The 3rd patch introduces DAMON module, it's data structures, and data structure
related common functions.  Following four patches (4nd to 7th) implements the
core mechanisms of DAMON, namely regions based sampling (patch 4), adaptive
regions adjustment (patches 5-6), and dynamic memory mapping chage adoption
(patch 7).

Following four patches are for low level users of DAMON.  The 8th patch
implements callbacks for each of monitoring steps so that users can do whatever
they want with the access patterns.  The 9th one implements recording of access
patterns in DAMON for better convenience and efficiency.  Each of next two
patches (10th and 11th) respectively adds a debugfs interface for privileged
people and/or programs in user space, and a tracepoint for other tracepoints
supporting tracers such as perf.

Two patches for high level users of DAMON follows.  To provide a minimal
reference to the debugfs interface and for high level use/tests of the DAMON,
the next patch (12th) implements an user space tool.  The 13th patch adds a
document for administrators of DAMON.

Next two patches are for tests.  The 14th and 15th patches provide unit tests
(based on kunit) and user space tests (based on kselftest), respectively.

Finally, the last patch (16th) updates the MAINTAINERS file.


Patch History
=============

The most biggest change in this version is support of minimal region size,
which defaults to 'PAGE_SIZE'.  This change will reduce unnecessary region
splits and thus improve the quality of the output.  In a future, we will be
able to make this configurable for support of various access check primitives
such as PMUs.

Changes from v9
(https://lore.kernel.org/linux-mm/20200427120442.24179-1-sjpark@amazon.com/)
 - Split each region into 4 subregions if possible (Jonathan Cameraon)
 - Update kunit test for the split code change

Changes from v8
(https://lore.kernel.org/linux-mm/20200406130938.14066-1-sjpark@amazon.com/)
 - Make regions always aligned by minimal region size that can be changed
   (Stefan Nuernberger)
 - Store binary format version in the recording file (Stefan Nuernberger)
 - Use 'int' for pid instead of 'unsigned long' (Stefan Nuernberger)
 - Fix a race condition in damon thread termination (Stefan Nuernberger)
 - Optimize random value generation and recording (Stefan Nuernberger)
 - Clean up commit messages and comments (Stefan Nuernberger)
 - Clean up code (Stefan Nuernberger)
 - Use explicit signalling and 'do_exit()' for damon thread termination 
 - Add more typos to spelling.txt
 - Update the performance evaluation results
 - Describe future plans in the cover letter

Changes from v7
(https://lore.kernel.org/linux-mm/20200318112722.30143-1-sjpark@amazon.com/)
 - Cleanup variable names (Jonathan Cameron)
 - Split sampling address setup from access_check() (Jonathan Cameron)
 - Make sampling address to always locate in the region (Jonathan Cameron)
 - Make initial region's sampling addr to be old (Jonathan Cameron)
 - Split kdamond on/off function to seperate functions (Jonathan Cameron)
 - Fix wrong kernel doc comments (Jonathan Cameron)
 - Reset 'last_accessed' to false in kdamond_check_access() if necessary
 - Rebase on v5.6

Changes from v6
(https://lore.kernel.org/linux-mm/20200224123047.32506-1-sjpark@amazon.com/)
 - Wordsmith cover letter (Shakeel Butt)
 - Cleanup code and commit messages (Jonathan Cameron)
 - Avoid kthread_run() under spinlock critical section (Jonathan Cameron)
 - Use kthread_stop() (Jonathan Cameron)
 - Change tracepoint to trace regions (Jonathan Cameron)
 - Implement API from the beginning (Jonathan Cameron)
 - Fix typos (Jonathan Cameron)
 - Fix access checking to properly handle regions smaller than single page
   (Jonathan Cameron)
 - Add found typos to 'scripts/spelling.txt'
 - Add recent evaluation results including DAMON-based Operation Schemes

Changes from v5
(https://lore.kernel.org/linux-mm/20200217103110.30817-1-sjpark@amazon.com/)
 - Fix minor bugs (sampling, record attributes, debugfs and user space tool)
 - selftests: Add debugfs interface tests for the bugs
 - Modify the user space tool to use its self default values for parameters
 - Fix pmg huge page access check

Changes from v4
(https://lore.kernel.org/linux-mm/20200210144812.26845-1-sjpark@amazon.com/)
 - Add 'Reviewed-by' for the kunit tests patch (Brendan Higgins)
 - Make the unit test to depedns on 'DAMON=y' (Randy Dunlap and kbuild bot)
   Reported-by: kbuild test robot <lkp@intel.com>
 - Fix m68k module build issue
   Reported-by: kbuild test robot <lkp@intel.com>
 - Add selftests
 - Seperate patches for low level users from core logics for better reading
 - Clean up debugfs interface
 - Trivial nitpicks

Changes from v3
(https://lore.kernel.org/linux-mm/20200204062312.19913-1-sj38.park@gmail.com/)
 - Fix i386 build issue
   Reported-by: kbuild test robot <lkp@intel.com>
 - Increase the default size of the monitoring result buffer to 1 MiB
 - Fix misc bugs in debugfs interface

Changes from v2
(https://lore.kernel.org/linux-mm/20200128085742.14566-1-sjpark@amazon.com/)
 - Move MAINTAINERS changes to last commit (Brendan Higgins)
 - Add descriptions for kunittest: why not only entire mappings and what the 4
   input sets are trying to test (Brendan Higgins)
 - Remove 'kdamond_need_stop()' test (Brendan Higgins)
 - Discuss about the 'perf mem' and DAMON (Peter Zijlstra)
 - Make CV clearly say what it actually does (Peter Zijlstra)
 - Answer why new module (Qian Cai)
 - Diable DAMON by default (Randy Dunlap)
 - Change the interface: Seperate recording attributes
   (attrs, record, rules) and allow multiple kdamond instances
 - Implement kernel API interface

Changes from v1
(https://lore.kernel.org/linux-mm/20200120162757.32375-1-sjpark@amazon.com/)
 - Rebase on v5.5
 - Add a tracepoint for integration with other tracers (Kirill A. Shutemov)
 - document: Add more description for the user space tool (Brendan Higgins)
 - unittest: Improve readability (Brendan Higgins)
 - unittest: Use consistent name and helpers function (Brendan Higgins)
 - Update PG_Young to avoid reclaim logic interference (Yunjae Lee)

Changes from RFC
(https://lore.kernel.org/linux-mm/20200110131522.29964-1-sjpark@amazon.com/)
 - Specify an ambiguous plan of access pattern based mm optimizations
 - Support loadable module build
 - Cleanup code

SeongJae Park (16):
  scripts/spelling: Add a few more typos
  mm/page_ext: Export lookup_page_ext() to GPL modules
  mm: Introduce Data Access MONitor (DAMON)
  mm/damon: Implement region based sampling
  mm/damon: Adaptively adjust regions
  mm/damon: Split regions into 4 subregions if possible
  mm/damon: Apply dynamic memory mapping changes
  mm/damon: Implement callbacks
  mm/damon: Implement access pattern recording
  mm/damon: Add debugfs interface
  mm/damon: Add tracepoints
  tools: Add a minimal user-space tool for DAMON
  Documentation/admin-guide/mm: Add a document for DAMON
  mm/damon: Add kunit tests
  mm/damon: Add user space selftests
  MAINTAINERS: Update for DAMON

 .../admin-guide/mm/data_access_monitor.rst    |  428 +++++
 Documentation/admin-guide/mm/index.rst        |    1 +
 MAINTAINERS                                   |   12 +
 include/linux/damon.h                         |   78 +
 include/trace/events/damon.h                  |   43 +
 mm/Kconfig                                    |   23 +
 mm/Makefile                                   |    1 +
 mm/damon-test.h                               |  622 +++++++
 mm/damon.c                                    | 1504 +++++++++++++++++
 mm/page_ext.c                                 |    1 +
 scripts/spelling.txt                          |    8 +
 tools/damon/.gitignore                        |    1 +
 tools/damon/_dist.py                          |   36 +
 tools/damon/_recfile.py                       |   23 +
 tools/damon/bin2txt.py                        |   67 +
 tools/damon/damo                              |   37 +
 tools/damon/heats.py                          |  362 ++++
 tools/damon/nr_regions.py                     |   91 +
 tools/damon/record.py                         |  212 +++
 tools/damon/report.py                         |   45 +
 tools/damon/wss.py                            |   97 ++
 tools/testing/selftests/damon/Makefile        |    7 +
 .../selftests/damon/_chk_dependency.sh        |   28 +
 tools/testing/selftests/damon/_chk_record.py  |  108 ++
 .../testing/selftests/damon/debugfs_attrs.sh  |  139 ++
 .../testing/selftests/damon/debugfs_record.sh |   50 +
 26 files changed, 4024 insertions(+)
 create mode 100644 Documentation/admin-guide/mm/data_access_monitor.rst
 create mode 100644 include/linux/damon.h
 create mode 100644 include/trace/events/damon.h
 create mode 100644 mm/damon-test.h
 create mode 100644 mm/damon.c
 create mode 100644 tools/damon/.gitignore
 create mode 100644 tools/damon/_dist.py
 create mode 100644 tools/damon/_recfile.py
 create mode 100644 tools/damon/bin2txt.py
 create mode 100755 tools/damon/damo
 create mode 100644 tools/damon/heats.py
 create mode 100644 tools/damon/nr_regions.py
 create mode 100644 tools/damon/record.py
 create mode 100644 tools/damon/report.py
 create mode 100644 tools/damon/wss.py
 create mode 100644 tools/testing/selftests/damon/Makefile
 create mode 100644 tools/testing/selftests/damon/_chk_dependency.sh
 create mode 100644 tools/testing/selftests/damon/_chk_record.py
 create mode 100755 tools/testing/selftests/damon/debugfs_attrs.sh
 create mode 100755 tools/testing/selftests/damon/debugfs_record.sh