refactor: replace rayon with NUMA-aware PartitionRunner

Replaces `rayon` parallel iteration across index, rebuild, reindex, and select modules with a custom `PartitionRunner`. This introduces NUMA-aware task distribution with CPU pinning and round-robin scheduling, eliminating `Arc`, `Mutex`, and atomic synchronization primitives in favor of a flat, pre-spawned worker architecture. Error handling is simplified via `.map_err()` and the `?` operator, while progress bar updates are decoupled into dedicated callbacks.

src/obicompactvec/src/bitmatrix.rs

@@ -1,5 +1,5 @@
 use std::fs::{self, File};
-use std::io::{self, Write as _};
+use std::io::{self, BufWriter, Write as _};
 use std::path::{Path, PathBuf};
 
 use memmap2::Mmap;
@@ -230,30 +230,47 @@ impl PackedBitMatrix {
 
 /// Build `presence/matrix.pbmx` from existing `col_*.pbiv` files.
 pub fn pack_bit_matrix(dir: &Path) -> io::Result<()> {
+    let packed_path = dir.join("matrix.pbmx");
+    if packed_path.exists() {
+        // Matrix complete; remove any leftover column files from a killed cleanup.
+        if let Ok(meta) = MatrixMeta::load(dir) {
+            for c in 0..meta.n_cols { let _ = fs::remove_file(col_path(dir, c)); }
+            let _ = fs::remove_file(dir.join("meta.json"));
+        }
+        return Ok(());
+    }
+
     let meta = MatrixMeta::load(dir)?;
     let n_cols = meta.n_cols;
 
-    let col_files: Vec<Vec<u8>> = (0..n_cols)
+    // Compute offsets from file sizes — no column data loaded into RAM.
-        .map(|c| fs::read(col_path(dir, c)))
+    let col_sizes: Vec<u64> = (0..n_cols)
+        .map(|c| fs::metadata(col_path(dir, c)).map(|m| m.len()))
         .collect::<io::Result<_>>()?;
 
-    let header_size = PBMX_HEADER + n_cols * 8;
+    let header_size = (PBMX_HEADER + n_cols * 8) as u64;
     let mut col_offset = header_size;
     let mut offsets = Vec::with_capacity(n_cols);
-    for data in &col_files {
+    for &size in &col_sizes {
-        offsets.push(col_offset as u64);
+        offsets.push(col_offset);
-        col_offset += data.len();
+        col_offset += size;
     }
 
-    let packed_path = dir.join("matrix.pbmx");
+    // Write to a temp file; rename atomically so a killed process never leaves
-    let mut file = File::create(&packed_path)?;
+    // a truncated matrix.pbmx that would be mistaken for a complete file.
-    file.write_all(&PBMX_MAGIC)?;
+    let tmp_path = dir.join("matrix.pbmx.tmp");
-    file.write_all(&[0u8; 4])?;
+    let mut out = BufWriter::new(File::create(&tmp_path)?);
-    file.write_all(&(meta.n as u64).to_le_bytes())?;
+    out.write_all(&PBMX_MAGIC)?;
-    file.write_all(&(n_cols as u64).to_le_bytes())?;
+    out.write_all(&[0u8; 4])?;
-    for &off in &offsets { file.write_all(&off.to_le_bytes())?; }
+    out.write_all(&(meta.n as u64).to_le_bytes())?;
-    for data in &col_files { file.write_all(data)?; }
+    out.write_all(&(n_cols as u64).to_le_bytes())?;
-    drop(file);
+    for &off in &offsets { out.write_all(&off.to_le_bytes())?; }
+    for c in 0..n_cols {
+        io::copy(&mut File::open(col_path(dir, c))?, &mut out)?;
+    }
+    out.flush()?;
+    drop(out);
+    fs::rename(&tmp_path, &packed_path)?;
 
     for c in 0..n_cols { fs::remove_file(col_path(dir, c))?; }
     fs::remove_file(dir.join("meta.json"))?;

src/obicompactvec/src/intmatrix.rs

@@ -1,6 +1,6 @@
 use std::cmp::Ordering;
 use std::fs::{self, File};
-use std::io::{self, Write as _};
+use std::io::{self, BufWriter, Write as _};
 use std::path::{Path, PathBuf};
 
 use memmap2::Mmap;
@@ -354,30 +354,47 @@ impl PackedCompactIntMatrix {
 
 /// Build `counts/matrix.pcmx` from existing `col_*.pciv` files.
 pub fn pack_compact_int_matrix(dir: &Path) -> io::Result<()> {
+    let packed_path = dir.join("matrix.pcmx");
+    if packed_path.exists() {
+        // Matrix complete; remove any leftover column files from a killed cleanup.
+        if let Ok(meta) = MatrixMeta::load(dir) {
+            for c in 0..meta.n_cols { let _ = fs::remove_file(col_path(dir, c)); }
+            let _ = fs::remove_file(dir.join("meta.json"));
+        }
+        return Ok(());
+    }
+
     let meta = MatrixMeta::load(dir)?;
     let n_cols = meta.n_cols;
 
-    let col_files: Vec<Vec<u8>> = (0..n_cols)
+    // Compute offsets from file sizes — no column data loaded into RAM.
-        .map(|c| fs::read(col_path(dir, c)))
+    let col_sizes: Vec<u64> = (0..n_cols)
+        .map(|c| fs::metadata(col_path(dir, c)).map(|m| m.len()))
         .collect::<io::Result<_>>()?;
 
-    let header_size = PCMX_HEADER + n_cols * 8;
+    let header_size = (PCMX_HEADER + n_cols * 8) as u64;
     let mut col_offset = header_size;
     let mut offsets = Vec::with_capacity(n_cols);
-    for data in &col_files {
+    for &size in &col_sizes {
-        offsets.push(col_offset as u64);
+        offsets.push(col_offset);
-        col_offset += data.len();
+        col_offset += size;
     }
 
-    let packed_path = dir.join("matrix.pcmx");
+    // Write to a temp file; rename atomically so a killed process never leaves
-    let mut file = File::create(&packed_path)?;
+    // a truncated matrix.pcmx that would be mistaken for a complete file.
-    file.write_all(&PCMX_MAGIC)?;
+    let tmp_path = dir.join("matrix.pcmx.tmp");
-    file.write_all(&[0u8; 4])?;
+    let mut out = BufWriter::new(File::create(&tmp_path)?);
-    file.write_all(&(meta.n as u64).to_le_bytes())?;
+    out.write_all(&PCMX_MAGIC)?;
-    file.write_all(&(n_cols as u64).to_le_bytes())?;
+    out.write_all(&[0u8; 4])?;
-    for &off in &offsets { file.write_all(&off.to_le_bytes())?; }
+    out.write_all(&(meta.n as u64).to_le_bytes())?;
-    for data in &col_files { file.write_all(data)?; }
+    out.write_all(&(n_cols as u64).to_le_bytes())?;
-    drop(file);
+    for &off in &offsets { out.write_all(&off.to_le_bytes())?; }
+    for c in 0..n_cols {
+        io::copy(&mut File::open(col_path(dir, c))?, &mut out)?;
+    }
+    out.flush()?;
+    drop(out);
+    fs::rename(&tmp_path, &packed_path)?;
 
     for c in 0..n_cols { fs::remove_file(col_path(dir, c))?; }
     fs::remove_file(dir.join("meta.json"))?;

src/obikindex/src/index.rs

@@ -1,8 +1,6 @@
 use std::collections::BTreeMap;
 use std::fs;
 use std::path::{Path, PathBuf};
-use std::sync::atomic::{AtomicUsize, Ordering};
-use std::sync::{Arc, Mutex};
 
 use obikpartitionner::{KmerPartition, KmerSpectrum};
 use obilayeredmap;
@@ -152,31 +150,25 @@ impl KmerIndex {
         let with_counts = self.meta.config.with_counts;
         let evidence = self.meta.config.evidence.clone();
         let block_bits = self.meta.config.block_bits;
-        let total_kmers = AtomicUsize::new(0);
+        let mut total_kmers: usize = 0;
+        let pb = progress_bar("index", n as u64, "partitions");
 
-        let pb = Arc::new(Mutex::new(progress_bar("index", n as u64, "partitions")));
+        let order: Vec<usize> = (0..n).collect();
-
+        let runner = crate::numa::PartitionRunner::new();
-        (0..n).into_par_iter().for_each(|i| {
+        runner.run(
-            match self.partition.build_index_layer(i, min_ab, max_ab, with_counts, &evidence, block_bits) {
+            &order,
-                Ok(0) => {}
+            |i| self.partition.build_index_layer(i, min_ab, max_ab, with_counts, &evidence, block_bits),
-                Ok(n_kmers) => {
+            |i, n_kmers, _| {
-                    total_kmers.fetch_add(n_kmers, Ordering::Relaxed);
+                if n_kmers > 0 {
-                    let pb = pb.lock().unwrap();
+                    total_kmers += n_kmers;
                     pb.inc(1);
                     pb.set_message(format!("{i}: {n_kmers} kmers"));
                 }
-                Err(e) => {
+            },
-                    eprintln!("error building layer for partition {i}: {e}");
+        ).map_err(OKIError::Partition)?;
-                    std::process::exit(1);
-                }
-            }
-        });
 
-        pb.lock().unwrap().finish_and_clear();
+        pb.finish_and_clear();
-        info!(
+        info!("done — {} total kmers indexed", total_kmers);
-            "done — {} total kmers indexed",
-            total_kmers.load(Ordering::Relaxed)
-        );
 
         if !keep_intermediate {
             for i in 0..n {
@@ -211,36 +203,25 @@ impl KmerIndex {
         use obilayeredmap::meta::PartitionMeta;
 
         let n = self.n_partitions();
-        let errors: Vec<_> = (0..n)
+        let order: Vec<usize> = (0..n).collect();
-            .into_par_iter()
+        crate::numa::PartitionRunner::new().run(
-            .filter_map(|i| {
+            &order,
+            |i| -> OKIResult<()> {
                 let index_dir = self.partition.part_dir(i).join("index");
-                if !index_dir.exists() { return None; }
+                if !index_dir.exists() { return Ok(()); }
-                let meta = match PartitionMeta::load(&index_dir) {
+                let meta = PartitionMeta::load(&index_dir)
-                    Ok(m) => m,
+                    .map_err(|e| OKIError::Io(std::io::Error::new(std::io::ErrorKind::Other, e.to_string())))?;
-                    Err(e) => return Some(OKIError::Io(std::io::Error::new(std::io::ErrorKind::Other, e.to_string()))),
-                };
                 for l in 0..meta.n_layers {
                     let layer_dir = index_dir.join(format!("layer_{l}"));
                     let presence_dir = layer_dir.join("presence");
                     let counts_dir   = layer_dir.join("counts");
-                    if presence_dir.exists() {
+                    if presence_dir.exists() { pack_bit_matrix(&presence_dir).map_err(OKIError::Io)?; }
-                        if let Err(e) = pack_bit_matrix(&presence_dir) {
+                    if counts_dir.exists()   { pack_compact_int_matrix(&counts_dir).map_err(OKIError::Io)?; }
-                            return Some(OKIError::Io(e));
-                        }
-                    }
-                    if counts_dir.exists() {
-                        if let Err(e) = pack_compact_int_matrix(&counts_dir) {
-                            return Some(OKIError::Io(e));
-                        }
-                    }
                 }
-                None
+                Ok(())
-            })
+            },
-            .collect();
+            |_, _, _| {},
-
+        )
-        if let Some(e) = errors.into_iter().next() { return Err(e); }
-        Ok(())
     }
 
     /// Write a `layer_meta.json` in any layer directory that is missing one.

src/obikindex/src/numa.rs

@@ -1,4 +1,4 @@
-// NUMA-aware Rayon thread pools via hwlocality.
+// NUMA-aware partition runner via hwlocality.
 //
 // Detects NUMA topology using hwloc (cross-platform: Linux, macOS, etc.) and
 // builds one Rayon ThreadPool per NUMA node with threads pinned to that node's
@@ -10,15 +10,14 @@
 //   - the system has only one NUMA node (UMA, Apple Silicon, single-socket)
 //   - any per-node pool fails to build
 
-use std::sync::{Arc, Mutex};
+use std::sync::Arc;
 use std::time::{Duration, Instant};
 
-use crossbeam_channel::{RecvTimeoutError, unbounded};
+use crossbeam_channel::unbounded;
 use hwlocality::Topology;
 use hwlocality::cpu::binding::CpuBindingFlags;
 use hwlocality::cpu::cpuset::CpuSet;
 use hwlocality::object::types::ObjectType;
-use obisys::CpuSample;
 use tracing::debug;
 
 // ── Public interface ──────────────────────────────────────────────────────────
@@ -104,27 +103,6 @@ fn build_pool(cpus: &[usize]) -> Option<rayon::ThreadPool> {
         .ok()
 }
 
-// ── Adaptive spawn heuristic ──────────────────────────────────────────────────
-//
-// First worker: spawn if CPU efficiency is below SPAWN_THRESHOLD (machine is
-// under-utilised). Subsequent workers: spawn only if the last worker raised
-// efficiency by at least the expected marginal gain (1/n_workers), with a
-// minimum floor to avoid spurious spawns from efficiency fluctuations.
-
-const SPAWN_THRESHOLD: f64 = 0.95;
-const MIN_MARGINAL_GAIN: f64 = 0.03;
-const SPAWN_POLL: Duration = Duration::from_secs(20);
-
-fn should_spawn_worker(n_workers: usize, eff: f64, eff_at_last_spawn: f64) -> bool {
-    if n_workers == 1 {
-        eff < SPAWN_THRESHOLD
-    } else {
-        let gain = eff - eff_at_last_spawn;
-        let expected = 1.0 / n_workers as f64;
-        gain >= (expected * 0.25).max(MIN_MARGINAL_GAIN)
-    }
-}
-
 // ── PartitionRunner ───────────────────────────────────────────────────────────
 
 struct NodeConfig {
@@ -135,42 +113,38 @@ struct NodeConfig {
 
 /// Generic NUMA-aware runner for partition-level parallel work.
 ///
-/// Encapsulates worker spawning, NUMA pinning, adaptive activation, and result
+/// Workers are distributed round-robin across NUMA nodes and pinned to their
-/// collection.  UMA systems are handled as the degenerate case of a single node
+/// node's CPUs.  UMA systems are the degenerate case: one node, no pinning.
-/// with no pinning.
 ///
-/// # Model
+/// # Termination
 ///
-/// One controller thread per NUMA node (one total on UMA).  Each controller
+/// Termination is driven entirely by channel closure:
-/// manages up to `max_workers` dormant workers that drain a shared work queue.
+///
-/// Workers are activated one at a time; a new worker is added when global CPU
+/// ```text
-/// efficiency justifies it.  On NUMA all workers are activated immediately
+/// drop(part_tx)    → part_rx drains → workers exit → drop their result_tx
-/// (memory bandwidth, not CPU count, is the bottleneck).
+/// drop(result_tx)  → result_rx closes → controller loop exits
+/// ```
+///
+/// No explicit counter or sentinel needed.
 pub struct PartitionRunner {
-    nodes:   Vec<NodeConfig>,
+    nodes: Vec<NodeConfig>,
-    n_cores: usize,
 }
 
 impl PartitionRunner {
-    /// Detect topology and build.  Falls back to a single-node UMA runner on
+    /// Total worker slots across all nodes.
-    /// macOS, single-socket machines, or hwloc failure.
-    /// Total number of pre-spawned worker slots across all nodes.
     pub fn max_workers(&self) -> usize {
         self.nodes.iter().map(|n| n.max_workers).sum()
     }
 
+    /// Detect topology and build.  Falls back to a single-node UMA runner on
+    /// macOS, single-socket machines, or hwloc failure.
     pub fn new() -> Self {
-        let n_cores = std::thread::available_parallelism()
-            .map(|n| n.get())
-            .unwrap_or(1);
-
         match build() {
             Some(ns) => {
                 let wpn = ns.workers_per_node();
                 debug!(
                     "PartitionRunner: NUMA mode — {} node(s) × {} worker(s)/node",
-                    ns.pools.len(),
+                    ns.pools.len(), wpn,
-                    wpn,
                 );
                 let nodes = ns.pools
                     .into_iter()
@@ -181,21 +155,20 @@ impl PartitionRunner {
                         max_workers: wpn,
                     })
                     .collect();
-                Self { nodes, n_cores }
+                Self { nodes }
             }
             None => {
+                let n_cores = std::thread::available_parallelism()
+                    .map(|n| n.get())
+                    .unwrap_or(1);
                 let max_workers = (n_cores / 2).max(1);
-                debug!(
+                debug!("PartitionRunner: UMA mode — {} worker(s)", max_workers);
-                    "PartitionRunner: UMA mode — adaptive up to {} worker(s)",
-                    max_workers,
-                );
                 Self {
                     nodes: vec![NodeConfig {
                         pool:        None,
                         cpu_ids:     vec![],
                         max_workers,
                     }],
-                    n_cores,
                 }
             }
         }
@@ -203,19 +176,17 @@ impl PartitionRunner {
 
     /// Run `f(i)` for every index in `order`.
     ///
-    /// `on_done(i, result, elapsed)` is called under an internal mutex as each
+    /// Workers are spawned upfront and distributed round-robin across NUMA
-    /// partition completes — suitable for progress bars, logging, and result
+    /// nodes.  `on_done(i, result, elapsed)` is called from the controller
-    /// aggregation.  No `Send` or `Sync` bound is required on the callback.
+    /// thread as each partition completes — suitable for progress bars and
-    ///
+    /// result aggregation.
-    /// The work queue is shared across all NUMA nodes: any idle worker takes
-    /// the next available partition regardless of node, ensuring load balance.
     ///
     /// Returns the first error produced by `f`, if any.
     pub fn run<F, R, E, C>(
         &self,
-        order:   &[usize],
+        order:      &[usize],
-        f:       F,
+        f:          F,
-        on_done: C,
+        mut on_done: C,
     ) -> Result<(), E>
     where
         F: Fn(usize) -> Result<R, E> + Send + Sync,
@@ -223,131 +194,56 @@ impl PartitionRunner {
         E: Send,
         C: FnMut(usize, R, Duration) + Send,
     {
-        let f        = Arc::new(f);
+        // Pre-load the work queue, then drop the sender so workers' part_rx
-        let on_done  = Arc::new(Mutex::new(on_done));
+        // iterators terminate when the queue is drained.
-        let first_err: Arc<Mutex<Option<E>>> = Arc::new(Mutex::new(None));
-
-        // Shared work queue — pre-loaded in caller-supplied order.
         let (part_tx, part_rx) = unbounded::<usize>();
-        for &i in order {
+        for &i in order { part_tx.send(i).ok(); }
-            part_tx.send(i).ok();
-        }
         drop(part_tx);
 
-        let n_cores = self.n_cores;
+        let (result_tx, result_rx) = unbounded::<(usize, Result<R, E>, Duration)>();
+        let n_nodes = self.nodes.len();
+        let f = &f; // shared borrow; F: Sync so concurrent calls are safe
+
+        let mut first_err: Option<E> = None;
 
         std::thread::scope(|s| {
-            for node in &self.nodes {
+            // Spawn all workers upfront, round-robin across NUMA nodes.
-                let f         = Arc::clone(&f);
+            for w in 0..self.max_workers() {
-                let on_done   = Arc::clone(&on_done);
+                let node    = &self.nodes[w % n_nodes];
-                let first_err = Arc::clone(&first_err);
+                let prx     = part_rx.clone();
-                let part_rx   = part_rx.clone();
+                let rtx     = result_tx.clone();
+                let pool    = node.pool.clone();
+                let cpu_ids = &node.cpu_ids;
 
                 s.spawn(move || {
-                    // Per-node result and activation channels.
+                    if !cpu_ids.is_empty() { pin_current_thread(cpu_ids); }
-                    let (result_tx, result_rx) =
+                    for i in &prx {
-                        unbounded::<(usize, Result<R, E>, Duration)>();
+                        let t = Instant::now();
-                    let (activate_tx, activate_rx) = unbounded::<()>();
+                        let r = match &pool {
-
+                            Some(p) => p.install(|| f(i)),
-                    std::thread::scope(|ws| {
+                            None    => f(i),
-                        // Pre-spawn workers (all dormant until activated).
+                        };
-                        for _ in 0..node.max_workers {
+                        rtx.send((i, r, t.elapsed())).ok();
-                            let prx = part_rx.clone();
+                    }
-                            let rtx = result_tx.clone();
-                            let arx = activate_rx.clone();
-                            let f   = Arc::clone(&f);
-                            let pool    = node.pool.clone();
-                            let cpu_ids = node.cpu_ids.clone();
-
-                            ws.spawn(move || {
-                                if !cpu_ids.is_empty() {
-                                    pin_current_thread(&cpu_ids);
-                                }
-                                if arx.recv().is_err() {
-                                    return; // never activated — exit cleanly
-                                }
-                                for i in &prx {
-                                    let t = Instant::now();
-                                    let r = match &pool {
-                                        Some(p) => p.install(|| f(i)),
-                                        None    => f(i),
-                                    };
-                                    rtx.send((i, r, t.elapsed())).ok();
-                                }
-                            });
-                        }
-                        // Drop the controller's copy: result_rx disconnects
-                        // once all worker copies are also dropped (workers done).
-                        drop(result_tx);
-
-                        // In NUMA mode activate all workers immediately;
-                        // in UMA mode activate one and grow adaptively.
-                        let numa_mode = node.pool.is_some();
-                        let initial   = if numa_mode { node.max_workers } else { 1 };
-                        for _ in 0..initial {
-                            activate_tx.send(()).ok();
-                        }
-                        let mut active_workers      = initial;
-                        let mut cpu_sample          = CpuSample::now();
-                        let mut eff_at_last_spawn   = 0.0f64;
-
-                        // Controller loop.
-                        loop {
-                            match result_rx.recv_timeout(SPAWN_POLL) {
-                                Ok((i, r, dur)) => {
-                                    match r {
-                                        Ok(v) => {
-                                            on_done.lock().unwrap()(i, v, dur);
-                                        }
-                                        Err(e) => {
-                                            let mut g = first_err.lock().unwrap();
-                                            if g.is_none() { *g = Some(e); }
-                                        }
-                                    }
-                                    if !numa_mode && active_workers < node.max_workers {
-                                        let eff = cpu_sample.cpu_efficiency(n_cores);
-                                        if should_spawn_worker(active_workers, eff, eff_at_last_spawn) {
-                                            debug!(
-                                                "activated worker {} — efficiency {:.0}%",
-                                                active_workers + 1,
-                                                eff * 100.0,
-                                            );
-                                            activate_tx.send(()).ok();
-                                            active_workers     += 1;
-                                            eff_at_last_spawn   = eff;
-                                            cpu_sample          = CpuSample::now();
-                                        }
-                                    }
-                                }
-                                Err(RecvTimeoutError::Timeout) => {
-                                    if !numa_mode && active_workers < node.max_workers {
-                                        let eff = cpu_sample.cpu_efficiency(n_cores);
-                                        if should_spawn_worker(active_workers, eff, eff_at_last_spawn) {
-                                            debug!(
-                                                "activated worker {} (poll) — efficiency {:.0}%",
-                                                active_workers + 1,
-                                                eff * 100.0,
-                                            );
-                                            activate_tx.send(()).ok();
-                                            active_workers     += 1;
-                                            eff_at_last_spawn   = eff;
-                                            cpu_sample          = CpuSample::now();
-                                        }
-                                    }
-                                }
-                                Err(RecvTimeoutError::Disconnected) => break,
-                            }
-                        }
-                        // Signal any dormant workers that were never activated
-                        // to exit (UMA mode where max_workers was never reached).
-                        drop(activate_tx);
-                    }); // ws: waits for all workers of this node
                 });
             }
-        }); // s: waits for all node controllers
 
-        let mut g = first_err.lock().unwrap();
+            // Drop the controller's sender: result_rx closes once all worker
-        match g.take() {
+            // rtx clones are dropped (i.e. all workers have exited).
+            drop(result_tx);
+
+            // Drain results concurrently with workers.  The for loop exits
+            // when result_rx is disconnected — at that point all workers are
+            // done and the scope join below is instantaneous.
+            for (i, r, dur) in &result_rx {
+                match r {
+                    Ok(v)  => on_done(i, v, dur),
+                    Err(e) => { if first_err.is_none() { first_err = Some(e); } }
+                }
+            }
+        });
+
+        match first_err {
             Some(e) => Err(e),
             None    => Ok(()),
         }

src/obikindex/src/rebuild.rs

@@ -4,7 +4,6 @@ use std::path::Path;
 
 use obikpartitionner::{KmerFilter, KmerPartition, MergeMode};
 use obisys::{Reporter, Stage, progress_bar};
-use rayon::prelude::*;
 use tracing::info;
 
 use crate::error::{OKIError, OKIResult};
@@ -83,23 +82,16 @@ impl KmerIndex {
         let src_partition = &src.partition;
         let block_bits = meta.config.block_bits;
 
-        let errors: Vec<obiskio::SKError> = (0..n_partitions)
+        let order: Vec<usize> = (0..n_partitions).collect();
-            .into_par_iter()
+        let runner = crate::numa::PartitionRunner::new();
-            .filter_map(|i| {
+        runner.run(
-                let result = dst_partition
+            &order,
-                    .rebuild_partition(src_partition, i, filters, mode, n_genomes, block_bits)
+            |i| dst_partition.rebuild_partition(src_partition, i, filters, mode, n_genomes, block_bits),
-                    .err();
+            |_, _, _| { pb.inc(1); },
-                pb.inc(1);
+        ).map_err(OKIError::Partition)?;
-                result
-            })
-            .collect();
 
         pb.finish_and_clear();
 
-        if let Some(e) = errors.into_iter().next() {
-            return Err(OKIError::Partition(e));
-        }
-
         rep.push(t.stop());
 
         // Write SENTINEL_INDEXED — output is ready to use.

src/obikindex/src/reindex.rs

@@ -3,7 +3,6 @@ use std::path::Path;
 use obilayeredmap::{IndexMode, layer::Layer};
 use obilayeredmap::meta::PartitionMeta;
 use obisys::{Reporter, Stage, progress_bar};
-use rayon::prelude::*;
 use tracing::info;
 
 use crate::error::{OKIError, OKIResult};
@@ -45,25 +44,17 @@ impl KmerIndex {
         let t = Stage::start("reindex");
         let pb = progress_bar("reindex", n as u64, "partitions");
 
-        let errors: Vec<String> = (0..n)
+        let order: Vec<usize> = (0..n).collect();
-            .into_par_iter()
+        let runner = crate::numa::PartitionRunner::new();
-            .filter_map(|i| {
+        runner.run(
-                let res = reindex_partition(
+            &order,
-                    &self.partition.part_dir(i).join("index"),
+            |i| reindex_partition(&self.partition.part_dir(i).join("index"), &target, block_bits)
-                    &target,
+                .map_err(|e| OKIError::InvalidInput(format!("partition {i}: {e}"))),
-                    block_bits,
+            |_, _, _| { pb.inc(1); },
-                );
+        )?;
-                pb.inc(1);
-                res.err().map(|e| format!("partition {i}: {e}"))
-            })
-            .collect();
 
         pb.finish_and_clear();
 
-        if let Some(e) = errors.into_iter().next() {
-            return Err(OKIError::InvalidInput(e));
-        }
-
         self.meta.config.evidence = target;
         if matches!(self.meta.config.evidence, IndexMode::Exact) {
             self.meta.config.block_bits = block_bits;

src/obikindex/src/select.rs

@@ -4,7 +4,6 @@ use std::path::Path;
 
 use obikpartitionner::{KmerPartition, OutputCol, PARTITIONS_SUBDIR};
 use obisys::{Stage, progress_bar};
-use rayon::prelude::*;
 use tracing::info;
 
 use crate::error::{OKIError, OKIResult};
@@ -72,25 +71,16 @@ impl KmerIndex {
         let pb  = progress_bar("select", n_partitions as u64, "partitions");
         let src_partition = &src.partition;
 
-        let errors: Vec<obiskio::SKError> = (0..n_partitions)
+        let order: Vec<usize> = (0..n_partitions).collect();
-            .into_par_iter()
+        let runner = crate::numa::PartitionRunner::new();
-            .filter_map(|i| {
+        runner.run(
-                let result = dst_partition.select_partition(
+            &order,
-                    src_partition, i, specs,
+            |i| dst_partition.select_partition(src_partition, i, specs, n_src_genomes, threshold, output_presence, false),
-                    n_src_genomes, threshold, output_presence,
+            |_, _, _| { pb.inc(1); },
-                    false,
+        ).map_err(OKIError::Partition)?;
-                );
-                pb.inc(1);
-                result.err()
-            })
-            .collect();
 
         pb.finish_and_clear();
 
-        if let Some(e) = errors.into_iter().next() {
-            return Err(OKIError::Partition(e));
-        }
-
         let _ = t.stop();
 
         fs::File::create(output.join(SENTINEL_INDEXED))?;
@@ -132,25 +122,17 @@ impl KmerIndex {
         let t  = Stage::start("select");
         let pb = progress_bar("select", n_partitions as u64, "partitions");
 
-        let errors: Vec<obiskio::SKError> = (0..n_partitions)
+        let partition = &self.partition;
-            .into_par_iter()
+        let order: Vec<usize> = (0..n_partitions).collect();
-            .filter_map(|i| {
+        let runner = crate::numa::PartitionRunner::new();
-                let result = self.partition.select_partition(
+        runner.run(
-                    &src_partition, i, specs,
+            &order,
-                    n_src_genomes, threshold, output_presence,
+            |i| partition.select_partition(&src_partition, i, specs, n_src_genomes, threshold, output_presence, true),
-                    true,
+            |_, _, _| { pb.inc(1); },
-                );
+        ).map_err(OKIError::Partition)?;
-                pb.inc(1);
-                result.err()
-            })
-            .collect();
 
         pb.finish_and_clear();
 
-        if let Some(e) = errors.into_iter().next() {
-            return Err(OKIError::Partition(e));
-        }
-
         let _ = t.stop();
 
         // Update index.meta with new genome list and with_counts flag.