A. Colbrook, E. Brewer, C. Dellarocas, and W. Weihl. An algorithm for concurrent search trees. In Proceedings of the 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....for main memory databases is discussed in [17] The paper addresses the problem of maintaining local copies of the centralized variables, and discusses various recovery mechanisms. Distributed memory data structures have been proposed by Ellis [3] Severance [17] Peleg [14] Colbrook et al. [2] and Johnson and Colbrook [6] Ellis [3] has proposed a distributed extendible hashing technique, that uses techniques similar to the ones we use here. Colbrook et al. 2] have proposed a pipelined distributed B tree, where each level of the tree is maintained by a different processor. The ....

....mechanisms. Distributed memory data structures have been proposed by Ellis [3] Severance [17] Peleg [14] Colbrook et al. 2] and Johnson and Colbrook [6] Ellis [3] has proposed a distributed extendible hashing technique, that uses techniques similar to the ones we use here. Colbrook et al. [2] have proposed a pipelined distributed B tree, where each level of the tree is maintained by a different processor. The parallelism achieved is limited by the height of the B tree and the processors are not data balanced. 2 Johnson and Colbrook [6] present a distributed B tree suitable for ....

Colbrook A., Brewer A. E., Dellarocas C.N. and Weihl E. W. An Algorithm for Concurrent Search Trees, Proccedings of the 20th International Conference on Parallel Processing, 1991, pp. 38-41.

....are completely local. A search operation examines one node at a time to find its key, and an insert operation searches for the node that contains its key, performs the insert, then restructures the tree from the bottom up. Some work has been done to develop a distributed B tree. Colbrook et al. [6] developed a pipelined algorithm. Wang and Weihl [39, 41] have proposed that parallel B trees be stored using Multi version Memory, a special cache coherence algorithm for linked data structures. Multi version Memory permits only a single update to occur on a replicated node at any point in time ....

A. Colbrook, E.A. Brewer, C.N. Dellarocas, and W.E. Weihl. An algorithm for concurrent search trees. In Proceedings of the 20th International Conference on Parallel Processing, pages III138--III141, 1991.

....are made locally thereby reducing the communication overhead and increasing parallelism. The paper also deals with the data balancing among processors. They suggest a way of reducing communication cost for data balancing by storing neighboring leaves on the same processor. Colbrook, et al. [6] have proposed a pipelined distributed B tree, where each level of the tree is maintained by a different processor. We used this approach as a preliminary design for our distributed B tree. The parallelism achieved is limited by the height of the B tree, and the processors are not data balanced ....

Colbrook A., Brewer A. E., Dellarocas C.N. and Weihl E. W. An Algorithm for Concurrent Search Trees, Proccedings of the 20th International Conference on Parallel Processing, 1991, pp. 38-41.

....are completely local. A search operation examines one node at a time to find its key, and an insert operation searches for the node that contains its key, performs the insert, then restructures the tree from the bottom up. Some work has been done to develop a distributed B tree. Colbrook et al. [3] developed a pipelined algorithm. Wang and Weihl [21] use a special form of cache coherence to implement a parallel B tree, so that it can be implemented on a shared nothing architecture with the appropriate underlying software. The dB tree [9, 10] implements the B link tree algorithm as a ....

A. Colbrook, E.A. Brewer, C.N. Dellarocas, and W.E. Weihl. An algorithm for concurrent search trees. In Proceedings of the 20th International Conference on Parallel Processing, pages III138--III141, 1991.

....at MIT [4, 5, 9] Proteus is an execution driven simulator that interleaves the execution of an application program with the simulation of the underlying architecture. This structure makes it possible for Proteus to achieve a high degree of accuracy, which has been confirmed by several experiments [7, 9]. We configured Proteus to run our experiments on k ary 1 cubes, 2 cubes, and 3 cubes of various sizes: 1, 2, 4, 8, 27, 64, and 125 processors. Although the Proteus simulator does allow us to simulate shared memory caches, we did not do so, as we did not want the choice of caching scheme to affect ....

Adrian Colbrook, Eric A. Brewer, Chrysanthos N. Dellarocas, and William E. Weihl. An Algo- rithm for Concurrent Search Trees. In Proceedings of the 1991.

....can be easily implemented. Peleg [26, 27] has proposed several structures for implementing a distributed dictionary. The concern of these papers is the message complexity of access and data balancing. However, the issues of efficiency and concurrent access are not addressed. Colbrook et al. [28] have proposed a pipelined distributed B tree, where each level of the tree is maintained by a different processor. The parallelism that can be obtained from this implementation is limited by the number of levels in the tree, and the distributed tree is not data balanced. The contribution of this ....

....gives up its copy of the node, it must elect a new birth processor, and distribute the decision to all owners on the node (a synchronizing operation) 2. 4 Parent Pointers Requiring every node to contain a pointer to its parent simplifies the task of finding parents and splitting the root [28, 17]. Parent pointers can be maintained in the following manner: When a node splits, some of the node s children are transferred to the new sibling. When a processor receives the split suboperation from the birth processor, it splits the node locally, then makes all locally managed children of the new ....

A. Colbrook, E.A. Brewer, C.N. Dellarocas, and W.E. Weihl. An algorithm for concurrent search trees. In Proceedings of the Oth International Conference on Parallel Processing, pages III138- III141, 1991.

....issues of concurrent restructuring. Deitzfelbinger and Meyer auf der Hyde [9] give algorithms for implementing a hash table on a synchronous network. Ranade [38] gives algorithms and performance bounds for implementing a search tree in a synchronous butterfly or mesh network. Colbrook et al. [8] have proposed a pipelined distributed B tree, where each level of the tree is maintained by a different processor. The parallelism that can be obtained from this implementation is limited by the number of levels in the tree, and the distributed tree is not data balanced. The contribution of this ....

A. Colbrook, E. Brewer, C. Dellarocas, and W. Weihl, An algorithm for concurrent search trees, in Proceedings of the 20th International Conference on Parallel Processing, 1991, pp. III138--III141.

....can be easily implemented. Peleg [26, 27] has proposed several structures for implementing a distributed dictionary. The concern of these papers is the message complexity of access and data balancing. However, the issues of efficiency and concurrent access are not addressed. Colbrook et al. [28] have proposed a pipelined distributed B tree, where each level of the tree is maintained by a different processor. The parallelism that can be obtained from this implementation is limited by the number of levels in the tree, and the distributed tree is not data balanced. The contribution of this ....

....gives up its copy of the node, it must elect a new birth processor, and distribute the decision to all owners on the node (a synchronizing operation) 2. 4 Parent Pointers Requiring every node to contain a pointer to its parent simplifies the task of finding parents and splitting the root [28, 17]. Parent pointers can be maintained in the following manner: When a node splits, some of the node s children are transferred to the new sibling. When a processor receives the split suboperation from the birth processor, it splits the node locally, then makes all locally managed children of the new ....

A. Colbrook, E.A. Brewer, C.N. Dellarocas, and W.E. Weihl. An algorithm for concurrent search trees. In Proceedings of the 20th International Conference on Parallel Processing, pages III138-- III141, 1991.

No context found.

A. Colbrook, E. Brewer, C. Dellarocas, and W. Weihl. An algorithm for concurrent search trees. In Proceedings of the 1991.

No context found.

Colbrook, A., Brewer, E. A., Dellarocas, C.N. and Weihl, W. E. An Algorithm for Concurrent Search Trees, Proccedings of the 20th International Conference on Parallel Processing, 1991, pp. 38-41.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC