1: \begin{abstract}

2: We study

3: an extension of the bin packing problem,

4: where packing together two or more items

5: may make them occupy less volume than the sum of their individual sizes.

6: To achieve this property,

7: an item is defined as a finite set of symbols from a given alphabet.

8: Unlike the items of \textsc{Bin Packing},

9: two such sets can share zero, one or more symbols.

10: The problem was first introduced

11: in 2011 by Sindelar et al. \cite{Sindelar:2011}

12: under the name of \textsc{VM Packing}

13: with the addition of hierarchical sharing constraints

14: making it suitable for virtual machine colocation.

15: Without these constraints,

16: we prefer the more general name of \textsc{Pagination}.

17: After formulating it as an integer linear program,

18: we try to approximate its solutions with several families of algorithms:

19: from straightforward adaptations of classical \textsc{Bin Packing} heuristics,

20: to dedicated algorithms (greedy and non-greedy),

21: to standard and grouping genetic algorithms.

22: All of them are studied first theoretically,

23: then experimentally on an extensive random test set.

24: Based upon these data,

25: we propose

26: a predictive measure of the \added[id=R2]{statistical} difficulty of a given instance,

27: and finally recommend which algorithm should be used in which case,

28: depending on either time constraints or quality requirements.

29: \end{abstract}

30: