B-trees with lazy parent split

Abstract

A B-tree variant that postpones parent node splittings due to upcoming items until a later access of the same node is examined. This technique aims to decrease the possibility of propagating splittings to upper levels so that more concurrency is achieved. Insertion and deletion algorithms are given. Time and space performance results are also reported and comparison with conventional B-trees is carried out. It is shown that this technique substantially improves the performance of small degree B-trees so that, indeed, concurrency is enhanced.

Fringe Analysis of 2-3 Trees with Lazy Parent Split

“…The lps technique is used to decrease the number of node splits during an insert operation; therefore, the number of locks is decreased and as a consequence enhanced concurrency is achieved. It has been shown that B-trees with lps achieve better storage performance [6]. However, a theoretical proof of this remark was lacking from the literature.…”

“…In [6] insertion and deletion algorithms have been given. When inserting a new key into a node containing less than 2d keys, it is placed among the other keys.…”

“…A B-tree with lazy parent split (lps) is a B-tree variant aiming at improving concurrency [5,6,7,8], which is an important parameter in database management systems and file organizations for multi-user environments. The lps technique is used to decrease the number of node splits during an insert operation; therefore, the number of locks is decreased and as a consequence enhanced concurrency is achieved.…”

A new approach on indexing mobile objects on the plane

“…Our construction is based on the use of a B-tree variant, which is called Lazy B-tree and has better dynamic performance as well as optimal I/O complexities for both searching and update operations [19]. An orthogonal effort towards developing yet another B-tree variant under the same name has been proposed in [24]. The Lazy B-tree of [19] is a simple but non-trivial externalization of the techniques introduced in [31].…”

Indexing Mobile Objects on the Plane Revisited

“…Our construction is based on the use of a Btree variant, which is called Lazy B-tree and has better dynamic performance as well as optimal I/O complexities for both searching and update operations [15]. An orthogonal effort towards developing another yet B-tree variant under the same name has been proposed in [20]. The Lazy B-tree of [15] is a simple but non-trivial externalization of the techniques introduced in [24].…”