|
|
|
|
|
|
by rp1
1679 days ago
|
|
|
I imagine the reason is that the max load on a single server is O(log log number of servers), but it’s not possible for all the servers to be at max load. For instance, imagine load balancing via random assignment. The theoretical max load of a server is receiving every request, but if one server receives more requests, then the other servers receive less. |
|
|