how_to_run_tpch.txt

  How to run TPC-H benchmark queries with alenka
  
  Get Alenka :
  https://github.com/antonmks/Alenka
  Get modernGPU :
  https://github.com/moderngpu/moderngpu
  
  Unpack alenka into a directory, unpack moderngpu into the same directory : /home/alenkatest/alenka/moderngpu
  Unpack tpch.zip tools into a directory, compile and run dbgen utility to create data files : 
  make 
  ./dbgen -s 1
  
  Copy created *.tbl files into alenka directory
  
  Run load scripts to load the data into alenka data files :
  unzip tpch_load_scripts.zip
  ./alenka load_customer.sql
  ./alenka load_lineitem.sql
  ./alenka load_nation.sql
  ./alenka load_orders.sql
  ./alenka load_part.sql
  ./alenka load_partsupp.sql
  ./alenka load_region.sql
  ./alenka load_supplier.sql
  
  If you created data files using large scale parameter or your gpu has insufficient memory you can specify a segment size smaller than default 1GB :
  ./alenka - l 500 load_lineitem.sql - this would load data at 500MB segments
  
  Run the queries :
  unzip tpch_queries.zip
  ./alenka -v q1.sql
  ...
  ./alenka -v q7.sql
  
  The results of queries are either in specified file ( STORE D INTO 'mytest.txt' USING ('|'); )
  or shown on the screen ( DISPLAY F2 USING ('|'); )