Changes based on sampling

Cargo.lock

@@ -4,9 +4,9 @@ version = 3
 
 [[package]]
 name = "ahash"
-version = "0.8.11"
+version = "0.8.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e89da841a80418a9b391ebaea17f5c112ffaaa96f621d2c285b5174da76b9011"
+checksum = "42cd52102d3df161c77a887b608d7a4897d7cc112886a9537b738a887a03aaff"
 dependencies = [
  "cfg-if",
  "getrandom",
@@ -101,9 +101,9 @@ dependencies = [
 
 [[package]]
 name = "bumpalo"
-version = "3.15.4"
+version = "3.16.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7ff69b9dd49fd426c69a0db9fc04dd934cdb6645ff000864d98f7e2af8830eaa"
+checksum = "79296716171880943b8470b5f8d03aa55eb2e645a4874bdbb28adb49162e012c"
 
 [[package]]
 name = "bumpalo-herd"

calculate/Cargo.toml

@@ -4,11 +4,13 @@ version = "0.1.0"
 edition = "2021"
 
 # See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+[profile.release]
+debug = true
 
 [dependencies]
 clap = { version = "4.5.2", features = ["derive"] }
 # This is the latest version that actually uses AES on arm64
-ahash = "0.8.11"
+ahash = "=0.8.8"
 bumpalo-herd = "0.1.2"
 anyhow = "1.0.80"
 indicatif = "0.17.8"

calculate/src/optimized.rs

@@ -24,15 +24,15 @@ struct Data {
     count: u64,
 }
 
-const PARALLELISM_FACTOR: usize = 2;
+const PARALLELISM_FACTOR: usize = 1;
 const BUFFER_SIZE: usize = 1 << 30;
 const UNIQUE_CITY_COUNT: usize = 10_000;
 
 pub(crate) fn run(args: Args) -> Result<()> {
-    let input = File::open(args.input)?;
+    let input_file = File::open(args.input)?;
     let mut mmap_options = MmapOptions::new();
     mmap_options.populate();
-    let input = unsafe { mmap_options.map(&input)? };
+    let input = unsafe { mmap_options.map(&input_file)? };
     input.advise(Advice::Sequential)?;
     input.advise(Advice::WillNeed)?;
 
@@ -164,8 +164,9 @@ fn chunk<'input, 'bump>(
     let sep = Chunk::splat(b';');
     let nl = Chunk::splat(b'\n');
     let mut pb_since_last_inc: usize = 0;
+    const PB_INC_THRESHOLD: usize = 100_000_000;
     'outer: while input.len() > LANES {
-        if pb_since_last_inc >= 10_000_000 {
+        if pb_since_last_inc >= PB_INC_THRESHOLD {
             pb.inc(pb_since_last_inc as u64);
             pb_since_last_inc = 0;
         }
@@ -238,7 +239,7 @@ fn chunk<'input, 'bump>(
         data.count += 1;
     }
     loop {
-        if pb_since_last_inc >= 10_000_000 {
+        if pb_since_last_inc >= PB_INC_THRESHOLD {
             pb.inc(pb_since_last_inc as u64);
             pb_since_last_inc = 0;
         }

flake.nix

@@ -41,6 +41,7 @@
         {package = pkgs.clangStdenv;}
         {package = pkgs.fenix.default.toolchain; }
         {package = pkgs.cargo-flamegraph;}
+        {package = pkgs.samply;}
         {package = pkgs.nil;}
         {package = pkgs.rust-analyzer-nightly;}
       ];