Ubiquitous B-Tree

  title={Ubiquitous B-Tree},
  author={Douglas Comer},
  journal={ACM Comput. Surv.},
  • Douglas Comer
  • Published 1 June 1979
  • Computer Science
  • ACM Comput. Surv.
B-trees have become, de facto, a standard for file organization. File indexes of users, dedicated database systems, and general-purpose access methods have all been proposed and nnplemented using B-trees This paper reviews B-trees and shows why they have been so successful It discusses the major variations of the B-tree, especially the B+-tree, contrasting the relatwe merits and costs of each implementatmn. It illustrates a general purpose access method whmh uses a B-tree. 

A compact B-tree

This paper describes a Patricia tree-based B-tree variant suitable for OLTP that has been implemented in iAnywhere ASA Version 8.0 and can provide significant space and performance benefits over existing ASA indexes.

B-Trees and Databases, Past and Future

  • R. Bayer
  • Computer Science
    Software Pioneers
  • 2002
UB-Trees, a new multidimensional variant of B-T trees for the next generation of databases is discussed, which has superb properties with respect to performance and scalability in multiuser environments.

B-Tree Traversals

The B-tree is a highly dynamic structure in which records are often moved from one page to another in structure modifications such as page splits caused by insertions and page merges caused by deletions.

Postorder B-tree construction

  • L. W. West
  • Computer Science
    ACM Southeast Regional Conference
  • 1992
A method is presented for building B-trees from a sorted list of records without node splitting, where the number of records stored per node can be independently specified for each level of the tree, what Knuth calls a generalized B-tree.

Sequential retrieval of B-trees and a file structure with a dense B-tree index

A simple fast method is given for sequentially retrieving all the records in a B-tree, and a file structure for database is proposed, which is easy to insert, delete or search a record, and also convenient to retrieve records in the sequential order of the keys.

Efficient Search of Multi-Dimensional B-Trees

A method whereby BTrees can beiciently used to retrieve small subsets, thus avoiding large scans of potentially huge tables, and the ability of this method to reduce the need for additional secondary indexes, thus saving space, maintenance cost, and random accesses is described.

Approach for Handling Duplicate Keys in B + Tree

An alternative approach to make the B~+ tree explicitly support the duplicate keys by revising the index structure is adopted, and the basic algorithms to implement the idea are provided.

An Improved Query Technique for Embedded Database Based on S-B Tree

A static balanced tree--S-B tree is proposed to substitute B-tree as index of embedded database SQLite of Fixed Assets Management System to reduce space waste by B+-tree index.

An empirical comparison of B-trees, compact B-trees and multiway trees

This paper examines general B-trees empirically and compares them with a less restrictive structure, the general multiway tree, and a more restrictive structure and indicates their relative utility for large and small data sets.

Efficient Search of Multidimensional B-Trees

A method whereby BTrees can beiciently used to retrieve small subsets, thus avoiding large scans of potentially huge tables, and the ability of this method to reduce the need for additional secondary indexes, thus saving space, maintenance cost, and random accesses is described.



B-trees re-examined

The B-tree and its variants have, with increasing frequency, been proposed as a basic storage structure for multiuser database applications. Here, three potential problems which must be dealt with in

Pagination of B*-trees with variable-length records

A strategy is presented for pagination of B*-trees with variable-length records that results in shallow trees with fast access times and the performance of this strategy in an application is presented, compared with that of another strategy, and analyzed.

Information retrieval: information storage and retrieval using AVL trees

The number of probes required to post items on or retrieve items from AVL trees are few enough to permit a suitably organized computer, using only conventional components, to keep up with the delivery of requests from a buffered hypertape.

Indexing Design Considerations

The structure and use of indexes that facuitate the retrieval and storage of records based on a specific value, value range, or value sequence of a given field of a record within one or more data sets are examined.

Identifier Search Mechanisms: A Survey and Generalized Model

A number of important techniques which are used to retrieve individually identified data records from a computer memory space are discussed, specifically, the areas of compiler construction and of data base management.

File Organization: On the Selection of Random Access Index Points for Sequential Files

An examination is made of the consequences of partially or completely replacing one or more index levels by linear interpolation procedures, for all possible configurations of the several types of key distributions investigated.

Optimal 2, 3-Trees

The 2,3-trees that are optimal in the sense of having minimal expected number of nodes visited per access are characterized in terms of their “profiles”. The characterization leads directly to a

A new representation for linear lists

A key idea is the construction of a number representation behaving as described above, which can be used to model the propagation of modifications in the B-tree along the finger path, with the advantage that access is cheap in the neighborhood of each finger.

The multics system: an examination of its structure

The author builds a picture of the life of a process in coexistence with other processes, and suggests ways to model or construct subsystems that are far more complex than could be implemented using predecessor computer facilities.

Hash Table Methods

It is shown that under certain conditmns the authors can guarantee that no two items belonging to a certain class will have the same hash code, thus providing an improvement over the usual notion of a hash code as a randomizing technique.