Re: QLC Performances

Sorry, the missing links:
Merge request with benchmarks:
https://gitlab.com/patatoid/ecto3_mnesia/-/merge_requests/56

Ecto Query to Qlc compiler:
https://gitlab.com/patatoid/ecto3_mnesia/-/blob/master/lib/ecto/adapters/mnesia/query/qlc.ex

Cheers,
Jean


Le mercredi 06 octobre 2021 à 14:37 +0200, Jean Parpaillon a écrit :
> Dear all,
> With Pascal in CC, I'm working on an elixir Ecto to mnesia adapter.
> 
> Ecto queries are represented as an AST, which is compiled into database
> specific calls:
> * `mnesia:read/2` for basic queries
> * Qlc for others
> 
> We are facing huge performance issues and have tried to optimize the
> library in many ways, but it looks like the performance bottleneck is
> now in Qlc itself.
> 
> The library now includes a benchmark. In this merge request, you will
> find a benchmark that fetches 1000 records randomly from a table, by a
> non-key column.
> 
> There a 3 implementations:
> * mnesia.get.int.idx -> `mnesia:index_read/3`
> * qlc.get.int.idx -> use hand written qlc query ([Rec || {'_', '_',
> IndexedIntField, '_', '_', '_', '_', '_'} = Rec <-
> mnesia:table('test_table'), IndexedIntField == B0].')
> * ecto.get.int.idx -> use Ecto adapter
> 
> Here is the result of the benchmark:
> 
> Name                         ips        average  deviation        
> median         99th %
> mnesia.get.int.idx         30.38       0.0329 s    ±21.16%       0.0308
> s       0.0655 s
> qlc.get.int.idx             0.62         1.61 s     ±0.81%         1.60
> s         1.63 s
> ecto.get.int.idx            0.34         2.94 s     ±1.90%         2.93
> s         3.02 s
> 
> Comparison: 
> mnesia.get.int.idx         30.38
> qlc.get.int.idx             0.62 - 48.82x slower +1.57 s
> ecto.get.int.idx            0.34 - 89.44x slower +2.91 s
> 
> Memory usage statistics:
> 
> Name                       average  deviation         median        
> 99th %
> mnesia.get.int.idx      0.00496 GB     ±0.10%     0.00496 GB    
> 0.00497 GB
> qlc.get.int.idx            1.71 GB     ±0.00%        1.71 GB       
> 1.71 GB
> ecto.get.int.idx           1.55 GB     ±0.00%        1.55 GB       
> 1.55 GB
> 
> Comparison: 
> mnesia.get.int.idx      0.00496 GB
> qlc.get.int.idx            1.71 GB - 344.04x memory usage +1.70 GB
> ecto.get.int.idx           1.55 GB - 313.04x memory usage +1.55 GB
> 
> We can observe a small overhead between plain qlc and Ecto, but a MAJOR
> performance drop between plain mnesia and qlc.
> 
> Has this performance drop been observed in other applications ? May I
> expect better performance with a different usage of qlc ?
> 
> I may try to fix qlc, would some bottlenecks have been identified or
> someone give me some hints on where those bottlenecks can be.
> 
> I have also had a look at erl_syntax module from syntax_tools. Does it
> seem a better approach to use erl_syntax for generating query handles,
> rather than existing qlc parse transform ?
> 
> I need and willing to help on this topic, and every hint is welcome.
> 
> Best regards,

-- 
Jean Parpaillon
--
Software Engineering Consultant
Chairman @ OW2 Consortium - https://ow2.org/
Président L'Insolite Compagnie - https://insolitecompagnie.com
--
Phone: +33 6 30 10 92 86
linkedin: http://www.linkedin.com/in/jeanparpaillon/en