Issue #12142 has been updated by Vladimir Makarov.


Yusuke Endoh wrote:
>
> It consumes more memory not only when they are small.
> 
> ~~~~
> # trunk
> $ /usr/bin/time -f %Mkb ./miniruby -e 'h = {}; 10000000.times {|n| h[n] = n }'
> 647512kb
> 
> # with your patch
> $ /usr/bin/time -f %Mkb ./miniruby -e 'h = {}; 10000000.times {|n| h[n] = n }'
> 793844kb
> ~~~~

I tried 4 different parameters for the test: 100000 1000000 10000000 100000000

The trunk ruby gives on my machine

15576kb
73308kb
650104kb
6520868kb

ruby with the proposed hash tables gives

15260kb
58268kb
795764kb
6300612kb

In 3 cases out of 4, the proposed hash tables are more compact than
the trunk ones.  So you were unfortunate with the test parameter.

It says for me the size of big tables is probably not an important problem.

But still sizes of small tables might be if it results in a slower code.
Unfortunately it does.  I confirm the slowdown problem exists for tables
of size <= 3 for your test in the first email.  So I will work on the
small table problem.

Doing performance improvements is a hard job (I've been experiencing it well doing such
job for GCC for last 20 years).  There are always tests where the result can
be worse.  The average results (or even better a geometric mean) on credible benchmarks should
be a major criterion.  If there is requirement that everything should
be better, there will be no progress on performance front after some
short initial time of the project.  I wrote this long paragraph here in order to make
other people expectation more clear -- whatever I'd do with the tables,
someone can find a test where the new tables will be worse on speed or/and
memory consumption.


----------------------------------------
Feature #12142: Hash tables with open addressing
https://bugs.ruby-lang.org/issues/12142#change-57297

* Author: Vladimir Makarov
* Status: Open
* Priority: Normal
* Assignee: 
----------------------------------------
~~~
 Hello, the following patch contains a new implementation of hash
tables (major files st.c and include/ruby/st.h).

  Modern processors have several levels of cache.  Usually,the CPU
reads one or a few lines of the cache from memory (or another level of
cache).  So CPU is much faster at reading data stored close to each
other.  The current implementation of Ruby hash tables does not fit
well to modern processor cache organization, which requires better
data locality for faster program speed.

The new hash table implementation achieves a better data locality
mainly by

  o switching to open addressing hash tables for access by keys.
    Removing hash collision lists lets us avoid *pointer chasing*, a
    common problem that produces bad data locality.  I see a tendency
    to move from chaining hash tables to open addressing hash tables
    due to their better fit to modern CPU memory organizations.
    CPython recently made such switch
    (https://hg.python.org/cpython/file/ff1938d12240/Objects/dictobject.c).
    PHP did this a bit earlier
    https://nikic.github.io/2014/12/22/PHPs-new-hashtable-implementation.html.
    GCC has widely-used such hash tables
    (https://gcc.gnu.org/svn/gcc/trunk/libiberty/hashtab.c) internally
    for more than 15 years.

  o removing doubly linked lists and putting the elements into an array
    for accessing to elements by their inclusion order.  That also
    removes pointer chaising on the doubly linked lists used for
    traversing elements by their inclusion order.

A more detailed description of the proposed implementation can be
found in the top comment of the file st.c.

The new implementation was benchmarked on 21 MRI hash table benchmarks
for two most widely used targets x86-64 (Intel 4.2GHz i7-4790K) and ARM
(Exynos 5410 - 1.6GHz Cortex-A15):

make benchmark-each ITEM=bm_hash OPTS='-r 3 -v' COMPARE_RUBY='<trunk ruby>'

Here the results for x86-64:

hash_aref_dsym       1.094
hash_aref_dsym_long          1.383
hash_aref_fix        1.048
hash_aref_flo        1.860
hash_aref_miss       1.107
hash_aref_str        1.107
hash_aref_sym        1.191
hash_aref_sym_long           1.113
hash_flatten         1.258
hash_ident_flo       1.627
hash_ident_num       1.045
hash_ident_obj       1.143
hash_ident_str       1.127
hash_ident_sym       1.152
hash_keys            2.714
hash_shift           2.209
hash_shift_u16       1.442
hash_shift_u24       1.413
hash_shift_u32       1.396
hash_to_proc         2.831
hash_values          2.701

The average performance improvement is more 50%.  ARM results are
analogous -- no any benchmark performance degradation and about the
same average improvement.

The patch can be seen as

https://github.com/vnmakarov/ruby/compare/trunk...hash_tables_with_open_addressing.patch

or in a less convenient way as pull request changes

https://github.com/ruby/ruby/pull/1264/files


This is my first patch for MRI and may be my proposal and
implementation have pitfalls.  But I am keen to learn and work on
inclusion of this code into MRI.

~~~



-- 
https://bugs.ruby-lang.org/

Unsubscribe: <mailto:ruby-core-request / ruby-lang.org?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>