tpc_h: Benchmark TPC-H queries

In ursa-labs/arrowbench: Tools for Continuous and Interactive Benchmarking

Benchmark TPC-H queries

Description

Benchmark TPC-H queries

Usage

tpc_h

Format

An object of class Benchmark of length 11.

Parameters

• engine One of c("arrow", "duckdb", "dplyr")

• query_id integer, 1-22

• format One of c("parquet", "feather", "native")

• scale_factor Scale factor to use for data generation (e.g. 0.1, 1, 10, 100)

• memory_map Should memory mapping be used when reading a file in? (only applicable to arrow, native. FALSE will result in the file being explicitly read into memory before the benchmark)

• output the format of the output (either "data_frame" (default) or "arrow_table")

• chunk_size a size of row groups to aim for in parquet or feather files (default: NULL is the default for arrow:write_parquet() or arrow::write_feather())

ursa-labs/arrowbench documentation built on July 8, 2023, 11:36 a.m.