[GitHub] [arrow-datafusion] yjshen commented on pull request #2132: WIP: Reduce sort memory usage v1

[GitBox <gi...@apache.org>]

yjshen commented on pull request #2132:
URL: https://github.com/apache/arrow-datafusion/pull/2132#issuecomment-1085799873


   A modified version of TPC-H q1:
   
   select
       l_returnflag,
       l_linestatus,
       l_quantity,
       l_extendedprice,
       l_discount,
       l_tax
   from
       lineitem
   order by
       l_extendedprice,
       l_discount;
   
   Before this PR:
   
   Running benchmarks with the following options: DataFusionBenchmarkOpt { query: 1, debug: false, iterations: 3, partitions: 2, batch_size: 8192, path: "/Users/shenyijie/code/arrow-datafusion/benchmarks/tpch-parquet", file_format: "parquet", mem_table: false, output_path: None }
   Query 1 iteration 0 took 5922.3 ms and returned 6001214 rows
   Query 1 iteration 1 took 5699.3 ms and returned 6001214 rows
   Query 1 iteration 2 took 6166.0 ms and returned 6001214 rows
   Query 1 avg time: 5929.20 ms
   
         
   With this PR:
                                                                              
   Running benchmarks with the following options: DataFusionBenchmarkOpt { query: 1, debug: false, iterations: 3, partitions: 2, batch_size: 8192, path: "/Users/shenyijie/code/arrow-datafusion/benchmarks/tpch-parquet", file_format: "parquet", mem_table: false, output_path: None }
   Query 1 iteration 0 took 4031.2 ms and returned 6001214 rows
   Query 1 iteration 1 took 4375.6 ms and returned 6001214 rows
   Query 1 iteration 2 took 4318.8 ms and returned 6001214 rows
   Query 1 avg time: 4241.89 ms
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org