Offline file assignments for online load balancing

We study a novel load balancing problem that arises in web search engines. The problem is a combination of an offline assignment problem, where files need to be (copied and) assigned to machines, and an online load balancing problem, where requests ask for specific files and need to be assigned to a corresponding machine, whose load is increased by this. We present simple deterministic algorithms for this problem and exhibit an interesting trade-off between the available space to make file copies and the obtainable makespan. We also give non-trivial lower bounds for a large class of deterministic algorithms and present a randomized algorithm that beats these bounds with high probability. Research highlights ► A novel file assignment and load balancing problem in web search engines is analyzed. ► Files are assigned to machines in an offline phase. ► In the following online phase machine load due to file accesses is balanced. ► Algorithms and lower bound for this problem are presented.