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
We connect to lower layer modules at random, choosing a small subset of modules to depend on (something like 5 / 100). As a result, the more modules there are in a layered graph, the more likely there are modules that would not be be connected to the main app graph since the amount of lower level modules that are connected to by higher level modules stays constant at 5 modules.
As a result %10-%20 of modules in most graphs are not connected to the main graph. Modules not connected to the main graph are not built by most build systems, so it looks like the layered graph is inherently faster than a flat graph, while it's only compiling %10-%20 less code & modules than a flat graph.
The text was updated successfully, but these errors were encountered:
We connect to lower layer modules at random, choosing a small subset of modules to depend on (something like 5 / 100). As a result, the more modules there are in a layered graph, the more likely there are modules that would not be be connected to the main app graph since the amount of lower level modules that are connected to by higher level modules stays constant at 5 modules.
As a result %10-%20 of modules in most graphs are not connected to the main graph. Modules not connected to the main graph are not built by most build systems, so it looks like the layered graph is inherently faster than a flat graph, while it's only compiling %10-%20 less code & modules than a flat graph.
The text was updated successfully, but these errors were encountered: