[PATCH v2 0/9] Introduce Data Access MONitor (DAMON)

From: sjpark
Date: Tue Jan 28 2020 - 03:58:34 EST


From: SeongJae Park <sjpark@xxxxxxxxx>

This patchset introduces a new kernel module for practical monitoring of data
accesses, namely DAMON.

The patches are organized in the following sequence. The first four patches
implements the core logic of DAMON one by one. After that, the fifth patch
implements DAMON's debugfs interface for users. To provide a minimal reference
to the low level interface and for more convenient use/tests of the DAMON, the
sixth patch implements an user space tool. The seventh patch adds a document
for administrators of DAMON, and the eightth patch provides DAMON's kunit
tests. Finally, the ninth patch implements a tracepoint for DAMON. As the
tracepoint prints every monitoring results, it will be easily integrated with
other tracers supporting tracepoints including perf.

The patches are based on the v5.5. You can also clone the complete git
tree:

$ git clone git://github.com/sjp38/linux -b damon/patches/v2

The web is also available:
https://github.com/sjp38/linux/releases/tag/damon/patches/v2


Patch History
-------------

Changes from v1
(https://lore.kernel.org/linux-mm/20200120162757.32375-1-sjpark@xxxxxxxxxx/)
- 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@xxxxxxxxxx/)
- Specify an ambiguous plan of access pattern based mm optimizations
- Support loadable module build
- Cleanup code

----

DAMON is a kernel module that allows users to monitor the actual memory access
pattern of specific user-space processes. It aims to be 1) accurate enough to
be useful for performance-centric domains, and 2) sufficiently light-weight so
that it can be applied online.

For the goals, DAMON utilizes its two core mechanisms, called region-based
sampling and adaptive regions adjustment. The region-based sampling allows
users to make their own trade-off between the quality and the overhead of the
monitoring and set the upperbound of the monitoring overhead. Further, the
adaptive regions adjustment mechanism makes DAMON to maximize the quality and
minimize the overhead with its best efforts while preserving the users
configured trade-off.

Please note that the term 'memory' in this document means 'main memory'. It
also assumes that it would usually utilizes the middle level speed memory
devices such as DRAMs or NVRAMs. CPU caches or storage devices are not our
concern, as those are too fast or too slow to be in DAMON's scope.


Background
==========

For performance-centric analysis and optimizations of memory management schemes
(either that of kernel space or user space), the actual data access pattern of
the workloads is highly useful. The information need to be only reasonable
rather than strictly correct, because some level of incorrectness can be
handled in many performance-centric domains. It also need to be taken within
reasonably short time with only light-weight overhead.

Manually extracting such data is not easy and time consuming if the target
workload is huge and complex, even for the developers of the programs. There
are a range of tools and techniques developed for general memory access
investigations, and some of those could be partially used for this purpose.
However, most of those are not practical or unscalable, mainly because those
are designed with no consideration about the trade-off between the accuracy of
the output and the overhead.

The memory access instrumentation techniques which is applied to many tools
such as Intel PIN is essential for correctness required cases such as invalid
memory access bug detections. However, those usually incur high overhead which
is unacceptable for many of the performance-centric domains. Periodic access
checks based on H/W or S/W access counting features (e.g., the Accessed bits of
PTEs or the PG_Idle flags of pages) can dramatically decrease the overhead by
forgiving some of the quality, compared to the instrumentation based
techniques. The reduced quality is still reasonable for many of the domains,
but the overhead can arbitrarily increase as the size of the target workload
grows. Miniature-like static region based sampling can set the upperbound of
the overhead, but it will now decrease the quality of the output as the size of
the workload grows.


Related Works
=============

There are a number of researches[1,2,3,4,5,6] optimizing memory management
mechanisms based on the actual memory access patterns that shows impressive
results. However, most of those has no deep consideration about the monitoring
of the accesses itself. Some of those focused on the overhead of the
monitoring, but does not consider the accuracy scalability[6] or has additional
dependencies[7]. Indeed, one recent research[5] about the proactive
reclamation has also proposed[8] to the kernel community but the monitoring
overhead was considered a main problem.

[1] Subramanya R Dulloor, Amitabha Roy, Zheguang Zhao, Narayanan Sundaram,
Nadathur Satish, Rajesh Sankaran, Jeff Jackson, and Karsten Schwan. 2016.
Data tiering in heterogeneous memory systems. In Proceedings of the 11th
European Conference on Computer Systems (EuroSys). ACM, 15.
[2] Youngjin Kwon, Hangchen Yu, Simon Peter, Christopher J Rossbach, and Emmett
Witchel. 2016. Coordinated and efficient huge page management with ingens.
In 12th USENIX Symposium on Operating Systems Design and Implementation
(OSDI). 705â721.
[3] Harald Servat, Antonio J PeÃa, GermÃn Llort, Estanislao Mercadal,
HansChristian Hoppe, and JesÃs Labarta. 2017. Automating the application
data placement in hybrid memory systems. In 2017 IEEE International
Conference on Cluster Computing (CLUSTER). IEEE, 126â136.
[4] Vlad Nitu, Boris Teabe, Alain Tchana, Canturk Isci, and Daniel Hagimont.
2018. Welcome to zombieland: practical and energy-efficient memory
disaggregation in a datacenter. In Proceedings of the 13th European
Conference on Computer Systems (EuroSys). ACM, 16.
[5] Andres Lagar-Cavilla, Junwhan Ahn, Suleiman Souhlal, Neha Agarwal, Radoslaw
Burny, Shakeel Butt, Jichuan Chang, Ashwin Chaugule, Nan Deng, Junaid
Shahid, Greg Thelen, Kamil Adam Yurtsever, Yu Zhao, and Parthasarathy
Ranganathan. 2019. Software-Defined Far Memory in Warehouse-Scale
Computers. In Proceedings of the 24th International Conference on
Architectural Support for Programming Languages and Operating Systems
(ASPLOS). ACM, New York, NY, USA, 317â330.
DOI:https://doi.org/10.1145/3297858.3304053
[6] Carl Waldspurger, Trausti Saemundsson, Irfan Ahmad, and Nohhyun Park.
2017. Cache Modeling and Optimization using Miniature Simulations. In 2017
USENIX Annual Technical Conference (ATC). USENIX Association, Santa
Clara, CA, 487â498.
https://www.usenix.org/conference/atc17/technical-sessions/
[7] Haojie Wang, Jidong Zhai, Xiongchao Tang, Bowen Yu, Xiaosong Ma, and
Wenguang Chen. 2018. Spindle: Informed Memory Access Monitoring. In 2018
USENIX Annual Technical Conference (ATC). USENIX Association, Boston, MA,
561â574. https://www.usenix.org/conference/atc18/presentation/wang-haojie
[8] Jonathan Corbet. 2019. Proactively reclaiming idle memory. (2019).
https://lwn.net/Articles/787611/.


Expected Use-cases
==================

A straightforward usecase of DAMON would be the program behavior analysis.
With the DAMON output, users can confirm whether the program is running as
intended or not. This will be useful for debuggings and tests of design
points.

The monitored results can also be useful for counting the dynamic working set
size of workloads. For the administration of memory overcommitted systems or
selection of the environments (e.g., containers providing different amount of
memory) for your workloads, this will be useful.

If you are a programmer, you can optimize your program by managing the memory
based on the actual data access pattern. For example, you can identify the
dynamic hotness of your data using DAMON and call ``mlock()`` to keep your hot
data in DRAM, or call ``madvise()`` with ``MADV_PAGEOUT`` to proactively
reclaim cold data. Even though your program is guaranteed to not encounter
memory pressure, you can still improve the performance by applying the DAMON
outputs for call of ``MADV_HUGEPAGE`` and ``MADV_NOHUGEPAGE``. More creative
optimizations would be possible. Our evaluations of DAMON includes a
straightforward optimization using the ``mlock()``. Please refer to the below
Evaluation section for more detail.

As DAMON incurs very low overhead, such optimizations can be applied not only
offline, but also online. Also, there is no reason to limit such optimizations
to the user space. Several parts of the kernel's memory management mechanisms
could be also optimized using DAMON. The reclamation, the THP (de)promotion
decisions, and the compaction would be such a candidates. Nevertheless,
current version of DAMON is not highly optimized for the online/in-kernel uses.


A Future Plan: Data Access Based Optimizations Support
------------------------------------------------------

As described in the above section, DAMON could be helpful for actual access
based memory management optimizations. Nevertheless, users who want to do such
optimizations should run DAMON, read the traced data (either online or
offline), analyze it, plan a new memory management scheme, and apply the new
scheme by themselves. It must be easier than the past, but could still require
some level of efforts. In its next development stage, DAMON will reduce some
of such efforts by allowing users to specify some access based memory
management rules for their specific processes.

Because this is just a plan, the specific interface is not fixed yet, but for
example, users will be allowed to write their desired memory management rules
to a special file in a DAMON specific format. The rules will be something like
'if a memory region of size in a range is keeping a range of hotness for more
than a duration, apply specific memory management rule using madvise() or
mlock() to the region'. For example, we can imagine rules like below:

# format is: <min/max size> <min/max frequency (0-99)> <duration> <action>

# if a region of a size keeps a very high access frequency for more than
# 100ms, lock the region in the main memory (call mlock()). But, if the
# region is larger than 500 MiB, skip it. The exception might be helpful
# if the system has only, say, 600 MiB of DRAM, a region of size larger
# than 600 MiB cannot be locked in the DRAM at all.
na 500M 90 99 100ms mlock

# if a region keeps a high access frequency for more than 100ms, put the
# region on the head of the LRU list (call madvise() with MADV_WILLNEED).
na na 80 90 100ms madv_willneed

# if a region keeps a low access frequency for more than 100ms, put the
# region on the tail of the LRU list (call madvise() with MADV_COLD).
na na 10 20 100ms madv_cold

# if a region keeps a very low access frequency for more than 100ms, swap
# out the region immediately (call madvise() with MADV_PAGEOUT).
na na 0 10 100ms madv_pageout

# if a region of a size bigger than 2MB keeps a very high access frequency
# for more than 100ms, let the region to use huge pages (call madvise()
# with MADV_HUGEPAGE).
2M na 90 99 100ms madv_hugepage

# If a regions of a size bigger than > 2MB keeps no high access frequency
# for more than 100ms, avoid the region from using huge pages (call
# madvise() with MADV_NOHUGEPAGE).
2M na 0 25 100ms madv_nohugepage


Mechanisms of DAMON
===================


Basic Access Check
------------------

DAMON basically reports what pages are how frequently accessed. The report is
passed to users in binary format via a ``result file`` which users can set it's
path. Note that the frequency is not an absolute number of accesses, but a
relative frequency among the pages of the target workloads.

Users can also control the resolution of the reports by setting two time
intervals, ``sampling interval`` and ``aggregation interval``. In detail,
DAMON checks access to each page per ``sampling interval``, aggregates the
results (counts the number of the accesses to each page), and reports the
aggregated results per ``aggregation interval``. For the access check of each
page, DAMON uses the Accessed bits of PTEs.

This is thus similar to the previously mentioned periodic access checks based
mechanisms, which overhead is increasing as the size of the target process
grows.


Region Based Sampling
---------------------

To avoid the unbounded increase of the overhead, DAMON groups a number of
adjacent pages that assumed to have same access frequencies into a region. As
long as the assumption (pages in a region have same access frequencies) is
kept, only one page in the region is required to be checked. Thus, for each
``sampling interval``, DAMON randomly picks one page in each region and clears
its Accessed bit. After one more ``sampling interval``, DAMON reads the
Accessed bit of the page and increases the access frequency of the region if
the bit has set meanwhile. Therefore, the monitoring overhead is controllable
by setting the number of regions. DAMON allows users to set the minimal and
maximum number of regions for the trade-off.

Except the assumption, this is almost same with the above-mentioned
miniature-like static region based sampling. In other words, this scheme
cannot preserve the quality of the output if the assumption is not guaranteed.


Adaptive Regions Adjustment
---------------------------

At the beginning of the monitoring, DAMON constructs the initial regions by
evenly splitting the memory mapped address space of the process into the
user-specified minimal number of regions. In this initial state, the
assumption is normally not kept and thus the quality could be low. To keep the
assumption as much as possible, DAMON adaptively merges and splits each region.
For each ``aggregation interval``, it compares the access frequencies of
adjacent regions and merges those if the frequency difference is small. Then,
after it reports and clears the aggregated access frequency of each region, it
splits each region into two regions if the total number of regions is smaller
than the half of the user-specified maximum number of regions.

In this way, DAMON provides its best-effort quality and minimal overhead while
keeping the bounds users set for their trade-off.


Applying Dynamic Memory Mappings
--------------------------------

Only a number of small parts in the super-huge virtual address space of the
processes is mapped to physical memory and accessed. Thus, tracking the
unmapped address regions is just wasteful. However, tracking every memory
mapping change might incur an overhead. For the reason, DAMON applies the
dynamic memory mapping changes to the tracking regions only for each of an
user-specified time interval (``regions update interval``).


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

A prototype of DAMON has evaluated on an Intel Xeon E7-8837 machine using 20
benchmarks that picked from SPEC CPU 2006, NAS, Tensorflow Benchmark,
SPLASH-2X, and PARSEC 3 benchmark suite. Nonethless, this section provides
only summary of the results. For more detail, please refer to the slides used
for the introduction of DAMON at the Linux Plumbers Conference 2019[1] or the
MIDDLEWARE'19 industrial track paper[2].


Quality
-------

We first traced and visualized the data access pattern of each workload. We
were able to confirm that the visualized results are reasonably accurate by
manually comparing those with the source code of the workloads.

To see the usefulness of the monitoring, we optimized 9 memory intensive
workloads among them for memory pressure situations using the DAMON outputs.
In detail, we identified frequently accessed memory regions in each workload
based on the DAMON results and protected them with ``mlock()`` system calls.
The optimized versions consistently show speedup (2.55x in best case, 1.65x in
average) under memory pressure situation.


Overhead
--------

We also measured the overhead of DAMON. It was not only under the upperbound
we set, but was much lower (0.6 percent of the bound in best case, 13.288
percent of the bound in average). This reduction of the overhead is mainly
resulted from the adaptive regions adjustment. We also compared the overhead
with that of the straightforward periodic Accessed bit check-based monitoring,
which checks the access of every page frame. DAMON's overhead was much smaller
than the straightforward mechanism by 94,242.42x in best case, 3,159.61x in
average.


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.

[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.

SeongJae Park (9):
mm: Introduce Data Access MONitor (DAMON)
mm/damon: Implement region based sampling
mm/damon: Adaptively adjust regions
mm/damon: Apply dynamic memory mapping changes
mm/damon: Add debugfs interface
mm/damon: Add minimal user-space tools
Documentation/admin-guide/mm: Add a document for DAMON
mm/damon: Add kunit tests
mm/damon: Add a tracepoint for result buffer writing

.../admin-guide/mm/data_access_monitor.rst | 401 +++++
Documentation/admin-guide/mm/index.rst | 1 +
MAINTAINERS | 10 +
include/trace/events/damon.h | 32 +
mm/Kconfig | 23 +
mm/Makefile | 1 +
mm/damon-test.h | 571 ++++++++
mm/damon.c | 1297 +++++++++++++++++
tools/damon/.gitignore | 1 +
tools/damon/_dist.py | 35 +
tools/damon/bin2txt.py | 64 +
tools/damon/damo | 37 +
tools/damon/heats.py | 358 +++++
tools/damon/nr_regions.py | 88 ++
tools/damon/record.py | 194 +++
tools/damon/report.py | 45 +
tools/damon/wss.py | 94 ++
17 files changed, 3252 insertions(+)
create mode 100644 Documentation/admin-guide/mm/data_access_monitor.rst
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/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

--
2.17.1