Merge pull request #14 from linksplatform/issue-13-23f296f3c251

Apply best practices from Comparisons.Neo4jVSDoublets

Author: konard

.github/workflows/rust.yml

@@ -1,6 +1,14 @@
 name: Benchmark Rust version
 
-on: [push, pull_request]
+on:
+  push:
+    branches:
+      - main
+      - master
+  pull_request:
+    branches:
+      - main
+      - master
 
 env:
   toolchain: nightly-2022-08-22
@@ -27,41 +35,98 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
+
       - name: Setup Rust
-        uses: actions-rs/toolchain@v1
+        uses: dtolnay/rust-toolchain@master
         with:
-          profile: minimal
           toolchain: ${{env.toolchain}}
           components: rustfmt, clippy
-          default: true
+
+      - name: Wait for PostgreSQL to be ready
+        run: |
+          echo "Waiting for PostgreSQL to be fully ready..."
+          for i in {1..30}; do
+            if pg_isready -h localhost -p 5432 -U postgres > /dev/null 2>&1; then
+              echo "PostgreSQL is ready!"
+              break
+            fi
+            echo "Attempt $i: PostgreSQL not ready yet..."
+            sleep 2
+          done
+
       - name: Build benchmark
-        uses: actions-rs/cargo@v1
-        with:
-          command: build
-          args: --release --all-features --manifest-path rust/Cargo.toml
+        run: cargo build --release --all-features --manifest-path rust/Cargo.toml
+
       - name: Run benchmark
         working-directory: rust
-        run: cargo bench --bench bench -- --output-format bencher | tee out.txt
+        env:
+          POSTGRES_HOST: localhost
+          POSTGRES_PORT: 5432
+          POSTGRES_USER: postgres
+          POSTGRES_PASSWORD: postgres
+          POSTGRES_DB: postgres
+          # Use 1000 links for main/master branch benchmarks, 10 for pull requests
+          BENCHMARK_LINKS: ${{ (github.event_name == 'push' && (github.ref == 'refs/heads/main' || github.ref == 'refs/heads/master')) && '1000' || '10' }}
+          # Use 100000 background links for main/master branch, 100 for pull requests
+          BENCHMARK_BACKGROUND_LINKS: ${{ (github.event_name == 'push' && (github.ref == 'refs/heads/main' || github.ref == 'refs/heads/master')) && '100000' || '100' }}
+        run: |
+          set -o pipefail
+          cargo bench --bench bench -- --output-format bencher | tee out.txt
+
       - name: Prepare benchmark results
+        if: github.event_name == 'push' && (github.ref == 'refs/heads/main' || github.ref == 'refs/heads/master')
         run: |
           git config --global user.email "[email protected]"
           git config --global user.name "LinksPlatformBencher"
           cd rust
           pip install numpy matplotlib
           python3 out.py
           cd ..
-          git fetch
-          git checkout gh-pages
-          mv -f rust/bench_rust.png Docs
-          mv -f rust/bench_rust_log_scale.png Docs
-          git add Docs
-          git commit -m "Publish new benchmark results"
-          git push origin gh-pages
+
+          # Create Docs directory if it doesn't exist
+          mkdir -p Docs
+
+          # Copy generated images
+          cp -f rust/bench_rust.png Docs/
+          cp -f rust/bench_rust_log_scale.png Docs/
+
+          # Update README with latest results
+          if [ -f rust/results.md ]; then
+            # Replace the results section in README.md
+            python3 -c "
+          import re
+
+          with open('rust/results.md', 'r') as f:
+              results = f.read()
+
+          with open('README.md', 'r') as f:
+              readme = f.read()
+
+          # Pattern to find and replace the results table
+          pattern = r'(\| Operation.*?\n\|[-|]+\n(?:\|.*?\n)*)'
+          if re.search(pattern, readme):
+              readme = re.sub(pattern, results.strip() + '\n', readme)
+
+          with open('README.md', 'w') as f:
+              f.write(readme)
+          "
+          fi
+
+          # Commit changes if any
+          git add Docs README.md
+          if git diff --staged --quiet; then
+            echo "No changes to commit"
+          else
+            git commit -m "Update benchmark results"
+            git push origin HEAD
+          fi
+
       - name: Save benchmark results
         uses: actions/upload-artifact@v4
         with:
           name: Benchmark results
           path: |
             rust/bench_rust.png
             rust/bench_rust_log_scale.png
+            rust/out.txt

Docs/.gitkeep

@@ -25,8 +25,8 @@ The results below represent the amount of time (ns) the operation takes per iter
 - Second picture shows time in a logarithmic scale (to see diffrence clearly, because it is around 2-3 orders of magnitude).
 
 ### Rust
-![Image of Rust benchmark (pixel scale)](https://github.com/linksplatform/Comparisons.PostgreSQLVSDoublets/blob/gh-pages/Docs/bench_rust.png?raw=true)
-![Image of Rust benchmark (log scale)](https://github.com/linksplatform/Comparisons.PostgreSQLVSDoublets/blob/gh-pages/Docs/bench_rust_log_scale.png?raw=true)
+![Image of Rust benchmark (pixel scale)](https://github.com/linksplatform/Comparisons.PostgreSQLVSDoublets/blob/main/Docs/bench_rust.png?raw=true)
+![Image of Rust benchmark (log scale)](https://github.com/linksplatform/Comparisons.PostgreSQLVSDoublets/blob/main/Docs/bench_rust_log_scale.png?raw=true)
 
 ### Raw benchmark results (all numbers are in nanoseconds)
 

README.md

@@ -1,8 +1,16 @@
 import re
+import logging
 import matplotlib.pyplot as plt
 import numpy as np
 
+# Enable detailed tracing. Set to False to disable verbose output.
+DEBUG = True
+logging.basicConfig(level=logging.INFO if DEBUG else logging.WARNING,
+                    format="%(message)s")
+
 data = open('out.txt').read()
+if DEBUG:
+    logging.info("Loaded out.txt, length: %d characters", len(data))
 
 patterns = [
     r"BM_(PSQL)/\w+\/?\w+?/(\w+)/\d+/min_warmup_time:20\.000\s+(\d+)\sns\s+\d+\sns\s+\d+",
@@ -18,15 +26,21 @@
 
 for pattern in patterns:
     matches = re.findall(pattern, data)
+    if DEBUG:
+        logging.info("Pattern %s matched %d entries", pattern, len(matches))
     for match in matches:
         if match[0] == 'PSQL':
             _category, transaction, time = match
+            if DEBUG:
+                logging.info("PSQL - %s: %s ns", transaction, time)
             if transaction == "Transaction":
                 PSQL_Transaction.append(int(time))
             else:
                 PSQL_NonTransaction.append(int(time))
         else:
             _category, trees, storage, time = match
+            if DEBUG:
+                logging.info("Doublets %s %s: %s ns", trees, storage, time)
             if trees == 'United':
                 if storage == 'Volatile':
                     Doublets_United_Volatile.append(int(time))
@@ -38,69 +52,92 @@
                 else:
                     Doublets_Split_NonVolatile.append(int(time))
 
+if DEBUG:
+    logging.info("\nFinal lists (after parsing):")
+    logging.info("PSQL_Transaction: %s", PSQL_Transaction)
+    logging.info("PSQL_NonTransaction: %s", PSQL_NonTransaction)
+    logging.info("Doublets_United_Volatile: %s", Doublets_United_Volatile)
+    logging.info("Doublets_United_NonVolatile: %s", Doublets_United_NonVolatile)
+    logging.info("Doublets_Split_Volatile: %s", Doublets_Split_Volatile)
+    logging.info("Doublets_Split_NonVolatile: %s", Doublets_Split_NonVolatile)
+
 labels = ['Create', 'Delete', 'Each Identity', 'Each Concrete', 'Each Outgoing', 'Each Incoming', 'Each All', 'Update']
+
+# ─────────────────────────────────────────────────────────────────────────────
+# Plots
+# ─────────────────────────────────────────────────────────────────────────────
+def ensure_min_visible(arr, min_val):
+    """Ensure non-zero values are at least min_val for visibility on graph."""
+    return [max(v, min_val) if v > 0 else 0 for v in arr]
+
 def bench1():
-    Doublets_United_Volatile_Pixels = [max(1, x // 10000000) for x in Doublets_United_Volatile]
-    Doublets_United_NonVolatile_Pixels = [max(1, x // 10000000) for x in Doublets_United_NonVolatile]
-    Doublets_Split_Volatile_Pixels = [max(1, x // 10000000) for x in Doublets_Split_Volatile]
-    Doublets_Split_NonVolatile_Pixels = [max(1, x // 10000000) for x in Doublets_Split_NonVolatile]
-    PSQL_NonTransaction_Pixels = [max(1, x // 10000000) for x in PSQL_NonTransaction]
-    PSQL_Transaction_Pixels = [max(1, x // 10000000) for x in PSQL_Transaction]
-    
+    """Horizontal bars – raw values (pixel scale)."""
     y = np.arange(len(labels))
-
     width = 0.1
-
     fig, ax = plt.subplots(figsize=(12, 8))
 
-
-    rects1 = ax.barh(y - 2*width, Doublets_United_Volatile_Pixels, width, label='Doublets United Volatile', color='salmon')
-    rects2 = ax.barh(y - width, Doublets_United_NonVolatile_Pixels, width, label='Doublets United NonVolatile', color='red')
-
-    rects3 = ax.barh(y, Doublets_Split_Volatile_Pixels, width, label='Doublets Split Volatile', color='lightgreen')
-    rects4 = ax.barh(y + width, Doublets_Split_NonVolatile_Pixels, width, label='Doublets Split NonVolatile', color='green')
-
-    rects5 = ax.barh(y + 2*width, PSQL_NonTransaction_Pixels, width, label='PSQL NonTransaction', color='lightblue')
-    rects6 = ax.barh(y + 3*width, PSQL_Transaction_Pixels, width, label='PSQL Transaction', color='blue')
-
-    ax.set_xlabel('Time (ns) - Scaled to Pixels')
-    ax.set_title('Benchmark comparison for Doublets and PostgreSQL')
+    # Calculate maximum value across all data series to determine scale
+    all_values = (Doublets_United_Volatile + Doublets_United_NonVolatile +
+                  Doublets_Split_Volatile + Doublets_Split_NonVolatile +
+                  PSQL_NonTransaction + PSQL_Transaction)
+    max_val = max(all_values) if all_values else 1
+
+    # Minimum visible bar width: ~0.5% of max value ensures at least 2 pixels
+    # on typical 12-inch wide figure at 100 DPI (~900px plot area)
+    min_visible = max_val * 0.005
+    if DEBUG:
+        logging.info("bench1: max_val=%d, min_visible=%d", max_val, min_visible)
+
+    # Apply minimum visibility to all data series
+    du_volatile_vis = ensure_min_visible(Doublets_United_Volatile, min_visible)
+    du_nonvolatile_vis = ensure_min_visible(Doublets_United_NonVolatile, min_visible)
+    ds_volatile_vis = ensure_min_visible(Doublets_Split_Volatile, min_visible)
+    ds_nonvolatile_vis = ensure_min_visible(Doublets_Split_NonVolatile, min_visible)
+    psql_non_vis = ensure_min_visible(PSQL_NonTransaction, min_visible)
+    psql_trans_vis = ensure_min_visible(PSQL_Transaction, min_visible)
+
+    ax.barh(y - 2*width, du_volatile_vis, width, label='Doublets United Volatile', color='salmon')
+    ax.barh(y - width, du_nonvolatile_vis, width, label='Doublets United NonVolatile', color='red')
+    ax.barh(y, ds_volatile_vis, width, label='Doublets Split Volatile', color='lightgreen')
+    ax.barh(y + width, ds_nonvolatile_vis, width, label='Doublets Split NonVolatile', color='green')
+    ax.barh(y + 2*width, psql_non_vis, width, label='PSQL NonTransaction', color='lightblue')
+    ax.barh(y + 3*width, psql_trans_vis, width, label='PSQL Transaction', color='blue')
+
+    ax.set_xlabel('Time (ns)')
+    ax.set_title('Benchmark Comparison: PostgreSQL vs Doublets (C++)')
     ax.set_yticks(y)
     ax.set_yticklabels(labels)
     ax.legend()
 
     fig.tight_layout()
-
-    plt.savefig("Bench1.png")
+    plt.savefig("bench_cpp.png")
     plt.close(fig)
+    if DEBUG: logging.info("bench_cpp.png saved.")
 
 def bench2():
+    """Horizontal bars – raw values on a log scale."""
     y = np.arange(len(labels))
-
     width = 0.1
     fig, ax = plt.subplots(figsize=(12, 8))
 
-    rects1 = ax.barh(y - 2*width, Doublets_United_Volatile, width, label='Doublets United Volatile', color='salmon')
-    rects2 = ax.barh(y - width, Doublets_United_NonVolatile, width, label='Doublets United NonVolatile', color='red')
+    ax.barh(y - 2*width, Doublets_United_Volatile, width, label='Doublets United Volatile', color='salmon')
+    ax.barh(y - width, Doublets_United_NonVolatile, width, label='Doublets United NonVolatile', color='red')
+    ax.barh(y, Doublets_Split_Volatile, width, label='Doublets Split Volatile', color='lightgreen')
+    ax.barh(y + width, Doublets_Split_NonVolatile, width, label='Doublets Split NonVolatile', color='green')
+    ax.barh(y + 2*width, PSQL_NonTransaction, width, label='PSQL NonTransaction', color='lightblue')
+    ax.barh(y + 3*width, PSQL_Transaction, width, label='PSQL Transaction', color='blue')
 
-    rects3 = ax.barh(y, Doublets_Split_Volatile, width, label='Doublets Split Volatile', color='lightgreen')
-    rects4 = ax.barh(y + width, Doublets_Split_NonVolatile, width, label='Doublets Split NonVolatile', color='green')
-
-    rects5 = ax.barh(y + 2*width, PSQL_NonTransaction, width, label='PSQL NonTransaction', color='lightblue')
-    rects6 = ax.barh(y + 3*width, PSQL_Transaction, width, label='PSQL Transaction', color='blue')
-
-    ax.set_xlabel('Time (ns) - Logarithmic Scale')
-    ax.set_title('Benchmark comparison for Doublets and PostgreSQL')
+    ax.set_xlabel('Time (ns) – log scale')
+    ax.set_title('Benchmark Comparison: PostgreSQL vs Doublets (C++)')
     ax.set_yticks(y)
     ax.set_yticklabels(labels)
     ax.legend()
-
     ax.set_xscale('log')
 
     fig.tight_layout()
-
-    plt.savefig("Bench2.png")
+    plt.savefig("bench_cpp_log_scale.png")
     plt.close(fig)
+    if DEBUG: logging.info("bench_cpp_log_scale.png saved.")
 
 bench1()
 bench2()

cpp/out.py

@@ -8,7 +8,7 @@ use {
         parts::LinkPart,
         split::{self, DataPart, IndexPart}, unit,
     },
-    linkspsql::{bench, Benched, Client, connect, Exclusive, Fork, Transaction},
+    linkspsql::{bench, benchmark_links, Benched, Client, connect, Exclusive, Fork, Transaction},
     std::{alloc::Global, time::{Duration, Instant}},
 };
 
@@ -17,9 +17,10 @@ fn bench<T: LinkType, B: Benched + Doublets<T>>(
     id: &str,
     mut benched: B,
 ) {
+    let links = benchmark_links();
     group.bench_function(id, |bencher| {
         bench!(|fork| as B {
-            for _ in 0..1_000 {
+            for _ in 0..links {
                 let _ = elapsed! {fork.create_point()?};
             }
         })(bencher, &mut benched);

rust/benches/benchmarks/create.rs

@@ -7,20 +7,22 @@ use {
         parts::LinkPart,
         split::{self, DataPart, IndexPart}, unit,
     },
-    linkspsql::{bench, Benched, Client, connect, Exclusive, Fork, Transaction, BACKGROUND_LINKS},
+    linkspsql::{bench, background_links, benchmark_links, Benched, Client, connect, Exclusive, Fork, Transaction},
     std::{alloc::Global, time::{Duration, Instant}},
 };
 fn bench<B: Benched + Doublets<usize>>(
     group: &mut BenchmarkGroup<WallTime>,
     id: &str,
     mut benched: B,
 ) {
+    let bg_links = background_links();
+    let links = benchmark_links();
     group.bench_function(id, |bencher| {
         bench!(|fork| as B {
-            for _prepare in BACKGROUND_LINKS..BACKGROUND_LINKS + 1_000 {
+            for _prepare in bg_links..bg_links + links {
                 let _ = fork.create_point();
             }
-            for id in (BACKGROUND_LINKS..=BACKGROUND_LINKS + 1_000).rev() {
+            for id in (bg_links..=bg_links + links).rev() {
                 let _ = elapsed! {fork.delete(id)?};
             }
         })(bencher, &mut benched);

rust/benches/benchmarks/delete.rs

@@ -8,7 +8,7 @@ use {
         split::{self, DataPart, IndexPart},
         unit, Doublets,
     },
-    linkspsql::{bench, connect, Benched, Client, Exclusive, Fork, Transaction},
+    linkspsql::{background_links, bench, connect, Benched, Client, Exclusive, Fork, Transaction},
     std::{
         alloc::Global,
         time::{Duration, Instant},
@@ -21,15 +21,10 @@ fn bench<B: Benched + Doublets<usize>>(
 ) {
     let handler = |_| Flow::Continue;
     let any = LinksConstants::new().any;
+    let bg_links = background_links();
     group.bench_function(id, |bencher| {
         bench!(|fork| as B {
-            for index in 1..=1_000 {
-                elapsed! {fork.each_by([any, index, index], handler)};
-            }
-            for index in 1_001..=2_000 {
-                elapsed! {fork.each_by([any, index, index], handler)};
-            }
-            for index in 2_001..=BACKGROUND_LINKS {
+            for index in 1..=bg_links {
                 elapsed! {fork.each_by([any, index, index], handler)};
             }
         })(bencher, &mut benched);