Skip to content

Use all cores for parquet->vortex conversion in duckdb-bench #9165

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
myrrc merged 1 commit into from
Aug 4, 2026
Merged

Use all cores for parquet->vortex conversion in duckdb-bench #9165

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 4 additions & 5 deletions vortex-bench/src/conversions.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,7 @@ use vortex::layout::layouts::compressed::CompressingStrategy;
use vortex::layout::layouts::flat::writer::FlatLayoutStrategy;
use vortex::session::VortexSession;
use vortex::utils::aliases::hash_set::HashSet;
use vortex::utils::parallelism::get_available_parallelism;
use vortex_arrow::FromArrowArray;
use vortex_arrow::FromArrowType;
use vortex_geo::extension::GeoMetadata;
Expand All @@ -72,9 +73,6 @@ const MEMORY_PER_STREAM_GB: u64 = 4;
/// Minimum number of concurrent conversion streams.
const MIN_CONCURRENCY: u64 = 1;

/// Maximum number of concurrent conversion streams. This is somewhat arbitary.
const MAX_CONCURRENCY: u64 = 16;

/// Returns the available system memory in bytes.
fn available_memory_bytes() -> u64 {
System::new_all().available_memory()
Expand All @@ -83,14 +81,15 @@ fn available_memory_bytes() -> u64 {
/// Calculate appropriate concurrency based on available memory.
fn calculate_concurrency() -> usize {
let available_gb = available_memory_bytes() / (1024 * 1024 * 1024);
let concurrency = (available_gb / MEMORY_PER_STREAM_GB).clamp(MIN_CONCURRENCY, MAX_CONCURRENCY);
let max_concurrency = get_available_parallelism().unwrap_or(1) as u64;
let concurrency = (available_gb / MEMORY_PER_STREAM_GB).clamp(MIN_CONCURRENCY, max_concurrency);

info!(
"Available memory: {}GB, maximum concurrency is: {}",
available_gb, concurrency
);

concurrency as usize
usize::try_from(concurrency).unwrap_or(1)
}

/// Read a Parquet file and return it as a Vortex [`ChunkedArray`].
Expand Down
Loading