We distinguish algorithm-level contributions related to graph problems from "runtime-level" concerns related to communication, scheduling, and other low-level features necessary to make distributed ...

In the decades that followed the algorithm’s publication, researchers tried to speed up the runtime by having the algorithm make more judicious choices — such as always using the shortest available ...

It combines techniques from Bellman-Ford, a slower algorithm and Dijkstra’s to improve runtime. It minimizies reliance on priority queue sorting by processing multiple nodes (the “frontier”) at once ...