• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11c
• /
• pp.1753-1756
• /
• 2002

주기억장치 상주형 색인구조에서는 2차 캐쉬 실패가 성능에 매우 큰 영향을 미친다. 기존에 제안된 주기억장치 상주형 색인구조들은 2차 캐쉬 실패를 고려하긴 했지만 여전히 트리의 각 레벨을 접근할 때는 2차 캐쉬실패가 발생한다. 본 논문에서는 이러한 문제점을 인식하고 트리 순회시 각 레벨을 방문할 때도 캐쉬 실패가 발생하지 않는 주기억장치 색인구조를 제안한다. 제안하는 색인구조는 다음 레벨에서 방문할 가능성이 있는 노드들을 프리페칭하여 다음 레벨을 방문할 때도 캐쉬 실패가 발생하지 않도록 한다. 또한, 기본적인 구조는 노드그룹 개념을 이용하여 노드의 팬-아웃을 증가시키는 CSB+-트리에 기반하지만 CSB+-트리의 다점인 분할 비용의 증가문제를 해결하기 위한 방법을 제안한다. 시뮬레이션을 통해 기존의 색인구조와 비교하여 제안하는 색인구조의 우수성을 보인다.

• Journal of KIISE:Information Networking
• /
• v.30 no.2
• /
• pp.282-292
• /
• 2003

This paper proposes an efficient data structure for forwarding IPv4 and IPv6 packets at the gigabit speed in backbone routers. The LPM(Longest Prefix Matching) search becomes a bottleneck of routers' performance since the LPM complexity grows in proportion to the forwarding table size and the address length. To speed up the forwarding process, this paper introduces a data structure named BMT(Bit-Map Tie) to minimize the frequent main memory accesses. All the necessary search computations in BMT are done over a small index table stored at cache. To build the small index table from the tie representation of the forwarding table, BMT represents a link pointer to the child node and a node pointer to the corresponding entry in the forwarding table with one bit respectively. To improve the poor performance of the conventional tries when their height becomes higher due to the increase of the address length, BMT adopts a binary search algorithm for determining the appropriate level of tries to start. The simulation experiments show that BMT compacts the IPv4 backbone routers' forwarding table into a small one less than 512-kbyte and achieves the average speed of 250ns/packet on Pentium II processors, which is almost the same performance as the fastest conventional lookup algorithms.

• The KIPS Transactions:PartD
• /
• v.10D no.1
• /
• pp.1-12
• /
• 2003

Recently, there have been many indexing schemes for multimedia data such as image, video data. But recent database applications, for example data mining and multimedia database, are required to support multi-user environment. In order for indexing schemes to be useful in multi-user environment, a concurrency control algorithm is required to handle it. So we propose a concurrency control algorithm that can be applied to CBF (cell-based filtering method), which uses the signature of the cell for alleviating the dimensional curse problem. In addition, we extend the SHORE storage system of Wisconsin university in order to handle high-dimensional data. This extended SHORE storage system provides conventional storage manager functions, guarantees the integrity of high-dimensional data and is flexible to the large scale of feature vectors for preventing the usage of large main memory. Finally, we implement the web-based image retrieval system by using the extended SHORE storage system. The key feature of this system is platform-independent access to the high-dimensional data as well as functionality of efficient content-based queries. Lastly. We evaluate an average response time of point query, range query and k-nearest query in terms of the number of threads.

• Spatial Information Research
• /
• v.18 no.4
• /
• pp.109-120
• /
• 2010

Recently, due to rapid increasement of spatial data collected from various geosensors in u-GIS environments, the importance of spatial index for efficient search of large spatial data is rising gradually. Especially, researches based R-Tree to improve search performance of spatial data have been actively performed. These previous researches focus on reducing overlaps between nodes or the height of the R -Tree. However, these can not solve an unnecessary node access problem efficiently occurred in tree traversal. In this paper, we propose a MR-Tree(Mapping-based R-Tree) to solve this problem and to support efficient search of large spatial data. The MR-Tree can improve search performance by using a mapping tree for direct access to leaf nodes of the R-Tree without tree traversal. The mapping tree is composed with MBRs and pointers of R-Tree leaf nodes associating each partition which is made by splitting data area repeatedly along dimensions. Especially, the MR-Tree can be adopted in various variations of the R-Tree easily without a modification of the R-Tree structure. In addition, because the mapping tree is constructed in main memory, search time can be greatly reduced. Finally, we proved superiority of MR-Tree performance through experiments.

• The KIPS Transactions:PartD
• /
• v.13D no.4 s.107
• /
• pp.463-476
• /
• 2006

R-trees have been traditionally optimized for the I/O performance with the disk page as the tree node. Recently, researchers have proposed cache-conscious variations of R-trees optimized for the CPU cache performance in main memory environments, where the node size is several cache lines wide and more entries are packed in a node by compressing MBR keys. However, because there is a big difference between the node sizes of two types of R-trees, disk-optimized R-trees show poor cache performance while cache-optimized R-trees exhibit poor disk performance. In this paper, we propose a cache and disk optimized R-tree, called the PR-tree (Prefetching R-tree). For the cache performance, the node size of the PR-tree is wider than a cache line, and the prefetch instruction is used to reduce the number of cache misses. For the I/O performance, the nodes of the PR-tree are fitted into one disk page. We represent the detailed analysis of cache misses for range queries, and enumerate all the reasonable in-page leaf and nonleaf node sizes, and heights of in-page trees to figure out tree parameters for best cache and I/O performance. The PR-tree that we propose achieves better cache performance than the disk-optimized R-tree: a factor of 3.5-15.1 improvement for one-by-one insertions, 6.5-15.1 improvement for deletions, 1.3-1.9 improvement for range queries, and 2.7-9.7 improvement for k-nearest neighbor queries. All experimental results do not show notable declines of the I/O performance.