feat: updated benchmarks

Author: domferr

include/process/base_process.hpp

@@ -201,6 +201,7 @@ class base_process {
 
         // parent
         PyOS_AfterFork_Parent();
+        LOGELAPSED("svc_fork time ", svc_init_fork);
         pickl.~pickling();
 
         // Release the main GIL

res.txt

@@ -1,18 +1 @@
-Running a farm of 1 workers and 5120 tasks. Each task is 50ms long and has a size of 100 bytes. Using processes-based strategy
-res[100] = res.get(100, []); res.get(100).append((1, 256227.337)) # bytes = 100 ms = 50
-Running a farm of 2 workers and 5120 tasks. Each task is 50ms long and has a size of 100 bytes. Using processes-based strategy
-res[100] = res.get(100, []); res.get(100).append((2, 128123.051)) # bytes = 100 ms = 50
-Running a farm of 4 workers and 5120 tasks. Each task is 50ms long and has a size of 100 bytes. Using processes-based strategy
-res[100] = res.get(100, []); res.get(100).append((4, 64074.72)) # bytes = 100 ms = 50
-Running a farm of 8 workers and 5120 tasks. Each task is 50ms long and has a size of 100 bytes. Using processes-based strategy
-res[100] = res.get(100, []); res.get(100).append((8, 32052.79)) # bytes = 100 ms = 50
-Running a farm of 12 workers and 5120 tasks. Each task is 50ms long and has a size of 100 bytes. Using processes-based strategy
-res[100] = res.get(100, []); res.get(100).append((12, 21393.738)) # bytes = 100 ms = 50
-Running a farm of 26 workers and 5120 tasks. Each task is 50ms long and has a size of 100 bytes. Using processes-based strategy
-res[100] = res.get(100, []); res.get(100).append((26, 9913.782)) # bytes = 100 ms = 50
-Running a farm of 36 workers and 5120 tasks. Each task is 50ms long and has a size of 100 bytes. Using processes-based strategy
-res[100] = res.get(100, []); res.get(100).append((36, 7215.942)) # bytes = 100 ms = 50
-Running a farm of 48 workers and 5120 tasks. Each task is 50ms long and has a size of 100 bytes. Using processes-based strategy
-res[100] = res.get(100, []); res.get(100).append((48, 5482.052)) # bytes = 100 ms = 50
-Running a farm of 64 workers and 5120 tasks. Each task is 50ms long and has a size of 100 bytes. Using processes-based strategy
-res[100] = res.get(100, []); res.get(100).append((64, 4197.795)) # bytes = 100 ms = 50
+25,28,29,30,32,30,34,38,36,38,40,38,44,43,42,43,44,46,47,50,47,51,44,47,47,45,

thesis/bench.ipynb

@@ -797,28 +797,28 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "[100, [256227.337, 128123.051, 64074.72, 32052.79, 21393.738, 9913.782, 7215.942, 5482.052, 4197.795]],\n"
+      "[10, [51252.803, 25647.669, 12847.717, 6462.909, 4367.318, 2109.714, 1616.25, 1304.495, 1086.843]],\n"
      ]
     }
    ],
    "source": [
     "res = dict()\n",
-    "res[100] = res.get(100, []); res.get(100).append((1, 256227.337)) # bytes = 100 ms = 50\n",
-    "res[100] = res.get(100, []); res.get(100).append((2, 128123.051)) # bytes = 100 ms = 50\n",
-    "res[100] = res.get(100, []); res.get(100).append((4, 64074.72)) # bytes = 100 ms = 50\n",
-    "res[100] = res.get(100, []); res.get(100).append((8, 32052.79)) # bytes = 100 ms = 50\n",
-    "res[100] = res.get(100, []); res.get(100).append((12, 21393.738)) # bytes = 100 ms = 50\n",
-    "res[100] = res.get(100, []); res.get(100).append((26, 9913.782)) # bytes = 100 ms = 50\n",
-    "res[100] = res.get(100, []); res.get(100).append((36, 7215.942)) # bytes = 100 ms = 50\n",
-    "res[100] = res.get(100, []); res.get(100).append((48, 5482.052)) # bytes = 100 ms = 50\n",
-    "res[100] = res.get(100, []); res.get(100).append((64, 4197.795)) # bytes = 100 ms = 50\n",
+    "res[10] = res.get(10, []); res.get(10).append((1, 51252.803)) # bytes = 10 ms = 50\n",
+    "res[10] = res.get(10, []); res.get(10).append((2, 25647.669)) # bytes = 10 ms = 50\n",
+    "res[10] = res.get(10, []); res.get(10).append((4, 12847.717)) # bytes = 10 ms = 50\n",
+    "res[10] = res.get(10, []); res.get(10).append((8, 6462.909)) # bytes = 10 ms = 50\n",
+    "res[10] = res.get(10, []); res.get(10).append((12, 4367.318)) # bytes = 10 ms = 50\n",
+    "res[10] = res.get(10, []); res.get(10).append((26, 2109.714)) # bytes = 10 ms = 50\n",
+    "res[10] = res.get(10, []); res.get(10).append((36, 1616.25)) # bytes = 10 ms = 50\n",
+    "res[10] = res.get(10, []); res.get(10).append((48, 1304.495)) # bytes = 10 ms = 50\n",
+    "res[10] = res.get(10, []); res.get(10).append((64, 1086.843)) # bytes = 10 ms = 50\n",
     "\n",
     "final_result = []\n",
     "for bytes, lis in res.items():\n",