Merge pull request #118 from Tmonster/update_benchmark_running_script

run version check all the time. also run large if machine is large

Author: Tmonster

.github/workflows/RunBenchmark.yml

@@ -93,6 +93,9 @@ jobs:
           if [ $ncores -eq 32 ]; then export MACHINE_TYPE="c6id.8xlarge"; fi
           if [ $ncores -eq 128 ]; then export MACHINE_TYPE="c6id.metal"; fi
           MACHINE_TYPE=$MACHINE_TYPE ./_run/run_small_medium.sh
+          if [ $ncores -eq 128 ]; then
+            MACHINE_TYPE=$MACHINE_TYPE ./_run/run_large.sh
+          fi
           
       - name: name new branch
         shell: bash

.github/workflows/regression.yml

@@ -19,7 +19,7 @@ jobs:
     matrix:
       solution: [data.table, collapse, dplyr, pandas, pydatatable, spark, juliadf, juliads, polars, R-arrow, duckdb, datafusion, dask, clickhouse]
   name: Solo solutions
-  runs-on: ubuntu-20.04
+  runs-on: ubuntu-latest
   env:
     CC: gcc-10
     CXX: g++-10

R-arrow/setup-R-arrow.sh

@@ -4,3 +4,5 @@ set -e
 # install stable arrow
 mkdir -p ./R-arrow/r-arrow
 Rscript -e 'install.packages(c("arrow","dplyr"), lib="./R-arrow/r-arrow")'
+
+./R-arrow/ver-R-arrow.sh

clickhouse/setup-clickhouse.sh

@@ -36,3 +36,5 @@ fi
 
 clickhouse-client --query "CREATE USER IF NOT EXISTS db_benchmark IDENTIFIED WITH no_password SETTINGS max_memory_usage = $MEMORY_LIMIT, max_bytes_before_external_group_by = $BYTES_BEFORE_EXTERNAL_GROUP_BY WRITABLE;"
 clickhouse-client --query "GRANT select, insert, create, alter, alter user, create table, truncate, drop, system flush logs on *.* to db_benchmark;"
+
+./clickhouse/ver-clickhouse.sh

collapse/setup-collapse.sh

@@ -4,3 +4,5 @@ set -e
 # install stable collapse
 mkdir -p ./collapse/r-collapse
 Rscript -e 'install.packages(c("Rcpp", "collapse"), lib="./collapse/r-collapse", repos = "http://cloud.r-project.org")'
+
+./collapse/ver-collapse.sh

dask/setup-dask.sh

@@ -15,3 +15,5 @@ python3 -m pip install "dask[complete]"
 # quit()
 
 deactivate
+
+./dask/ver-dask.sh

datatable/setup-datatable.sh

@@ -2,3 +2,5 @@
 # install devel data.table
 mkdir -p ./datatable/r-datatable
 Rscript -e 'install.packages("data.table", repos="https://Rdatatable.gitlab.io/data.table", method="curl", lib="./datatable/r-datatable")'
+
+./datatable/ver-datatable.sh

dplyr/setup-dplyr.sh

@@ -4,3 +4,5 @@ set -e
 # install stable dplyr
 mkdir -p ./dplyr/r-dplyr
 Rscript -e 'install.packages("dplyr", lib="./dplyr/r-dplyr", repos = "http://cloud.r-project.org")'
+
+./dplyr/ver-dplyr.sh

duckdb/setup-duckdb.sh

@@ -7,17 +7,6 @@ mkdir -p ./duckdb/r-duckdb
 # Rscript -e  'withr::with_libpaths(new = "./duckdb/r-duckdb", devtools::install_github("duckdb/duckdb/tools/rpkg"))'
 # prevent errors when running 'ver-duckdb.sh'
 Rscript -e 'install.packages("DBI", lib="./duckdb/r-duckdb", repos = "http://cloud.r-project.org")'
-
-
-cd duckdb
-rm -rf duckdb-r
-git clone https://github.com/duckdb/duckdb-r.git
-cd duckdb-r
-git checkout v1.2.0
-cd ..
-ncores=`python3 -c 'import multiprocessing as mp; print(mp.cpu_count())'`
-MAKE="make -j$ncores" R CMD INSTALL -l "./r-duckdb" duckdb-r
-rm -rf duckdb-r
-cd ..
+Rscript -e 'install.packages("duckdb", lib="./duckdb/r-duckdb", repos = "http://cloud.r-project.org")'
 
 ./duckdb/ver-duckdb.sh