You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
"\u001b[31mImportError\u001b[39m: cannot import name 'Edge' from 'asyncflow.components' (/home/gioele/projects/AsyncFlow/src/asyncflow/components/__init__.py)"
"> **Why small deltas appear:** warm-up effects, the user-sampling window (piecewise-constant rate), finite simulation horizon, and a (small) deterministic network latency naturally introduce small Theory vs Observed gaps. Increasing the simulation time and reducing network latency typically shrinks these deltas.\n"
303
+
"> **Why small deltas appear:** warm-up effects, the user-sampling window (piecewise-constant rate), finite simulation horizon. Increasing the simulation time typically shrinks these deltas.\n"
" Arrivals are produced by the same **two-stage, windowed Poisson sampler**: in each user-sampling window \\$\\Delta\\$, we draw the active users \\$U\\$ (Poisson or Normal, per config).\n",
95
95
" Within the window, arrivals are a **homogeneous Poisson process** with rate \\$\\Lambda = U \\cdot \\lambda\\_r/60\\$.\n",
96
96
"\n",
97
-
" With **small \\$\\Delta\\$**, **Poisson users**, **long runs**, and **tiny edge latency**, the aggregate arrivals seen by the load balancer approximate a global Poisson input, yielding a good empirical match to the M/M/c model.\n",
0 commit comments