Set union-find 問題是電腦科學的經典問題之一,從VLSI到多核心處理器的排程問題都和 set union-find 問題有關。 一些需要處理大量資訊的問題像是地理資訊系統也會用到 set union-find 的概念。
Set union-find 問題分為兩種,online 和 offline。不同的問題種類解決的策略也不同。
我們介紹不同模型下的 online 和 offline 問題。 對於 n 個元素執行 m 個混合的 union 或 find 指令,在 RAM model 下目前最快的 online 演算法時間複雜度為O(mα(m + n), n) + n)。 最快的 RAM model offline 演算法時間複雜度為O(m + n),由 Gabow 和 Tarjan 提出。
在 external memory model 下,我們關注的是資料在記憶體和外部儲存空間之間的輸入或輸出量。 在此模型下目前最快的 offline 演算法複雜度為O( (m+n)/B logM/B (m+n)/B ) I/Os。 而最快的 online 演算法是利用 Tarjan 和 Gabow 在 RAM model 提出的演算法,其複雜度為O(mα(m + n) + n) I/Os。
最後,我們也進行了 set union-find 問題在 parallel disk model (PDM) 上的研究,這目前是重要的開放問題。 如果要利用目前的技術解決這個問題,會有一些困難,我們找出了這個解決方向的瓶頸,可以進一步的研究解決。

Set union-find problem is a classical problem in computer science. From VLSI channel routing to multiprocessor scheduling can all be reduced to set union-find problems. In large scale implementations like geographic information system are also important issue today.
The set union-find problems has two types, online and offline. Different types of set union-find problems need different strategies to solve them.
We introduce the best-known algorithms of online and offline set union-find problems under RAM model and external memory model. For the algorithm that executes an intermixed sequence of m union and find operations on n elements, the fastest known algorithm for online version under RAM model runs in O(mα(m + n), n) + n) time. The best-known algorithm of offline version under RAM model, which is proposed by Gabow and Tarjan, runs in O(m + n) time.
In the external memory model, we measure the complexity by number of input and output between memory and the secondary storage. For the algorithm that executes an intermixed sequence of N not redundant union and find operations on set of elements, the optimal algorithm for offline external memory version runs in O( (m+n)/B logM/B (m+n)/B ) I/Os. In online external memory version, the current best result is by using Tarjan’s RAM model algorithm. The complexity is O(mα(m + n) + n) I/Os.
Finally, we propose the study of set union-find problem in the Parallel Disk Model (PDM) as an important open problem, and identify the bottlenecks of solving such a problem using the existing techniques.

[1] P. K. Agarwal, L. Arge, and K. Yi, “I/O-Efficient Batched Union-Find and Its Applications to Terrain Analysis”, in Proceedings of ACM Symposium on Computational Geometry, pp.167–176, 2006.
[2] A. Agarwal and J. S. Vitter, “The Input/Output Complexity of Sorting and Related Problems”, in Communications of the ACM, 31(9):1116–1127, 1988.
[3] T. H. Cormen, C. E. Leiserson, R. L. Rivest, and C. Stein, Introduction to Algorithms, MIT Press, 2001.
[4] H. N. Gabow and R. E. Tarjan, “A Linear Time Algorithm for a Special Case of Disjoint Set Union”, in Journal of Computer and System Sciences, 30(2):209–221, 1985. A preliminary version appeared in Proceedings of ACM Symposium on Theory of Computing, pp. 246–251, 1983.
[5] J. A. La Poutr ́e, “Lower Bounds for the Union-Find and the Split-Find Problem on Pointer Machines”, in Journal of Computer and System Sciences, 52(1):87–99, 1996. A preliminary version appeared in Proceedings of ACM Symposium on Theory of Computing, pp. 34–44, 1990.
[6] K. Mehlhorn, S. N ̈aher and H. Alt, “A Lower Bound on the Complexity of the Union-Split-Find Problem”, in SIAM Journal on Computing, 17(6):1093–1102, 1988.
[7] P. B. Miltersen, “Lower bouds for union-split-find related problems on randem access machines”, in Proceedings of ACM Symposium on Theory of Computing, pp. 625–634, 1994.
[8] R. E. Tarjan, “Efficiency of a Good But Not Linear Set Union Algorithm,” in Journal of the ACM, 22(2):215–225, 1975.
[9] J. S. Vitter and E. A. M. Shriver, “Optimal Disk I/O with Parallel Block Transfer”, in Proceedings of ACM Symposium on Theory of Computing, pp. 159–169, 1990.