load_compressed.sh

#!/usr/bin/env bash
set -euo pipefail

if [ $# -lt 1 ]; then
  echo "usage: $0 <benchmark> [benchmark args...]" >&2
  echo "  e.g. $0 tpch -s 10" >&2
  exit 1
fi

COMMON=(
  --umbradev
  --setting usedirectio=true
  --setting parallel=16
  --setting optimizer.disableindex=true
  --setting optimizer.sidewayinformationpassing=false
  --setting optimizer.forcedetokenize=false
  --setting relation.column.lazydecompress=false
  --setting relation.column.compressnumerics=false
)

# title | umbra_db | tokenize | compression
SYSTEMS=(
  "ASCII|ascii|false|None"
  "Token|token|true|None"
  "ASCII,OnPair|ascii_onpair|false|OnPair"
  "Token,OnPair|token_onpair|true|OnPair"
  "ASCII,Zstd|ascii_zstd|false|Zstd"
  "Token,Zstd|token_zstd|true|Zstd"
  "ASCII,LZ4|ascii_lz4|false|LZ4"
  "Token,LZ4|token_lz4|true|LZ4"
  "ASCII,FSST|ascii_fsst|false|FSST"
  "Token,FSST|token_fsst|true|FSST"
  "ASCII,OnPairBlock|ascii_onpairblock|false|OnPairBlock"
  "Token,OnPairBlock|token_onpairblock|true|OnPairBlock"
)

for s in "${SYSTEMS[@]}"; do
  IFS='|' read -r title db tok comp <<< "$s"
  echo "=== Loading Umbra ($title) into $* ==="
  python3 dbgen.py "${COMMON[@]}" \
    --parameter umbra_db="$db" \
    --setting tokenize="$tok" \
    --setting relation.column.compression="$comp" \
    "$@"
done