We consider an optimal labelling problem for a rooted directed tree abbreviated . as ''RDT'' which is motivated by certain scheduling problem. We obtain several necessary and sufficient conditions for the optimal labellings of a RDT and give a polynomially bounded algorithm for constructing the opti
Parallel Shortcutting of Rooted Trees
β Scribed by Mikkel Thorup
- Publisher
- Elsevier Science
- Year
- 1997
- Tongue
- English
- Weight
- 305 KB
- Volume
- 23
- Category
- Article
- ISSN
- 0196-6774
No coin nor oath required. For personal study only.
β¦ Synopsis
First it is shown that for any rooted tree T with n vertices, and parameter m G n, there is a ''shortcutting'' set S of at most m arcs from the transitive closure Ε½ . T* of T such for any Β¨, w g T *, there is a dipath in T j S from Β¨to w of length Ε½ Ε½ .. Ε½ O β£ m, n . An equivalent result has been achieved by Chazelle Algorithmica 2 Ε½ . . 1987 , 337α361 , but our proof is algorithmically simpler, and, in particular, it lends itself well to parallelization. More precisely, suppose that weights from a semigroup are assigned to the arcs of T. Then we can preprocess T in time Ε½ . Ε½ . Olog n with O mrlog n processors on a CREW PRAM such that for any Ε½ . Ε½ Ε½ .. Β¨,w gT*, we can find the weight of the path from Β¨to w in O β£ m, n Ε½ sequential time. Alon and Schieber ''Optimal Preprocessing for Answering On-. Line Product Queries,'' Technical Report 71r87, Tel Aviv University, 1987 have claimed that such a parallelization is possible for Chazelle's result. This claim is used in the optimal parallel sensitivity analysis for minimum spanning trees by Ε½ Dixon ''Minimum Spanning Tree Verification, Fast Priority Queues, and Mas-. sively Parallel Factoring,'' Ph.D. thesis, Princeton University, 1993 . However, Alon and Schieber did not give the details of the parallelization. Here we present a full proof, and our algorithms, both the sequential and the parallel versions, are rather simple, hence likely to be of practical relevance.
π SIMILAR VOLUMES
Denote by \(S_{n}\) the set of all distinct rooted binary trees with \(n\) unlabeled vertices. Define \(\sigma_{n}\) as a total height of a tree chosen at random in the set \(S_{n}\), assuming that all the possible choices are equally probable. The total height of a tree is defined as the sum of the
Trees are a useful data type, but they are not routinely included in parallel programming systems, in part because their irregular structure makes partitioning and scheduling difficult. We present a method for algebraically constructing implementations of tree skeletons, high-level homomorphic opera
An unbiased random generator for binary trees is developed for a CREW-PRAM. The generator is capable of generating a binary tree on \(n\) nodes in time \(O(\log n)\), space \(O(n)\), with \(O(n)\) processors; it is also capable of generating various related combinatorial objects. O 1994 Academic Pre
The quantification of the topological features of binary trees has been applied in several branches of biology, from botany to neurobiology to animal behaviour. The methods available for quantifying tree topology differ, both in how they are applied and how they relate to one another. In this paper,