Skyline query在近十年來受到廣泛的關注，其原因是因為它可以運用在很多應用上，尤其是在處理多決策問題時，但是在現在資訊爆炸的時代中，skyline query處理大量資料時，會非常耗費時間。對於處理大量數據的問題，Hadoop的MapReduce框架非常擅長處理數據密集型的問題，這篇論文中，我們使用MapReduce框架提出了一個角錐式分割方法，來處理skyline query問題。
在角錐式分割方法中，我們能夠使得數據量負載平衡，以確保在MapReduce平行處理的架構中，更快速的完成skyline query。

In recent years, skyline queries have received extensive attentions. It has many applications specialized on multi-criteria decision problems, such as recommender systems, confident search. The main approach for speeding up the processing of skyline queries is to filter out most of big data in the processing. Some researchers proposed grid-based partitioning method for data partitioning. However, it filters out less data, resulting in reducing its performance. Other researchers proposed the angle-based partitioning method for data partitioning, which can filter out more data. However, it cannot derive more partitions and needs much more computations for deriving the location of a data point, resulting in degrading its performance.
The aim of this proposal is to propose a pyramid-based processing of skyline queries in MapReduce for speeding up the skyline queries. The proposed pyramid-based partitioning method can easily partition the hypercube into many small pyramids and quickly identify the location of any data point. The local skyline points can be derived swiftly by using two-level BNL method. Furthermore, the dominated points can be found quickly by employing vertical-dominated method such t that the global skyline points are derived speedily. Thus this pyramid-based fashion of proposed algorithm can have highly parallel processing and filter out the data quickly, resulting in speeding up the system performance significantly.

[1] J. Lee, S. Hwang, Z. Nie, and J. Wen, "Navigation system for product search," in 2010 IEEE 26th International Conference on Data Engineering (ICDE 2010), 2010, pp. 1113-1116.
[2] T. Lappas and D. Gunopulos, "Efficient Confident Search in Large Review Corpora," in ECMLPKDD, 2010.
[3] L. Zou, L. Chen, M. T. Özsu, and D. Zhao, "Dynamic Skyline Queries in Large Graphs," in Database Systems for Advanced Applications, Berlin, Heidelberg, 2010, pp. 62-78: Springer Berlin Heidelberg.
[4] K. Mullesgaard, J. L. Pederseny, H. Lu, and Y. Zhou, "Efficient Skyline Computation in MapReduce," in EDBT, 2014.
[5] (2012). Big data meets big data analytics [Online]. Available: Available: http://www.sas.com/content/dam/SAS/en_us/doc/whitepaper1/bigdata-meets-big-data-analytics-105777.pdf
[6] Y. Park, J.-K. Min, and K. Shim, "Parallel computation of skyline and reverse skyline queries using mapreduce " Proc. VLDB Endow, vol. 6, no. 14, pp. 2002-2013, 2013.
[7] J. Zhang, X. Jiang, W.-S. Ku, and X. Qin, "Efficient Parallel Skyline Evaluation Using MapReduce," IEEE Transactions on Parallel and Distributed Systems, vol. 27, no. 7, pp. 1996-2009, 2016.
[8] A. Vlachou, C. Doulkeridis, and Y. Kotidis, Angle-based space partitioning for efficient parallel skyline computation. 2008, pp. 227-238.
[9] Y. Park, J.-K. Min, and K. Shim, "Efficient Processing of Skyline Queries Using MapReduce," IEEE Transactions on Knowledge and Data Engineering, vol. 29, no. 5, pp. 1031-1044, 2017.
[10] S. Borzsony, D. Kossmann, and K. Stocker, "The Skyline operator," in Proceedings 17th International Conference on Data Engineering, 2001, pp. 421-430.