Various communication aspects of locally distributed database processing are studied, using some distributed sorting and distributed hash-based join algorithms as examples. The algorithms are implemented on diskless w...
详细信息
Various communication aspects of locally distributed database processing are studied, using some distributed sorting and distributed hash-based join algorithms as examples. The algorithms are implemented on diskless workstations connected by an Ethernet network to simulate a distributed main memory system environment. This experimental testbed is described. Raw communication performance data (i.e. memory-to-memory data transfer timing) are presented. The effects of the underlying distributed operating system and the speed of the processor on the communication performance are shown. Two distributed sorting algorithms are used as examples to study the issue of concurrent transmissions of messages. distributed hash join is used as a case study for communication/local-processing tradeoff. The idea of load sharing among a number of sites to speed up the join operation is introduced.< >
A distributed control system for robotic applications is presented. This system allows a real-time performance increase, by parallelism of computation, and offers great modularity and versatility in robot control algo...
详细信息
A distributed control system for robotic applications is presented. This system allows a real-time performance increase, by parallelism of computation, and offers great modularity and versatility in robot control algorithm implementation. The robot control theory based on task-functions is outlined, and a model for robotic applications is proposed. This philosophy is more general than the classical hierarchical structure and allows smart sensor feedback at the servo. Software and hardware architectures are proposed to implement such control algorithms. All the system activity depends on communication between tasks and so the design of a real-time communication system is considered.< >
A heterogeneous network of parallel computers developed for complex distributed-processing applications is described. Network computers include a Connection Machine, a Butterfly multiprocessor, a WARP systolic array, ...
详细信息
A heterogeneous network of parallel computers developed for complex distributed-processing applications is described. Network computers include a Connection Machine, a Butterfly multiprocessor, a WARP systolic array, and a Symbolics and several SUN workstations. An Ethernet and a high-bandwidth APTEC bus support data transfers. distributed applications are built from individual processes executing on computers in the network. A powerful asynchronous communication facility is built upon the multiple computer operating systems to provide uniform message passing, global memory variables, and remote process execution services to processes. An executive controller and the LISP+ functional language provide a method of integrating distributed processes into an application with transparent control of network resources and communications. Additional applications can be rapidly built from existing processing to support experiments in distributed and parallel applications.< >
The original optimistic concurrency control scheme (H.T. Kung, and J.T. Robinson, 1979) has weaknesses with respect to the validation technique, the treatment of long transactions, and the problem of starvation. Some ...
详细信息
The original optimistic concurrency control scheme (H.T. Kung, and J.T. Robinson, 1979) has weaknesses with respect to the validation technique, the treatment of long transactions, and the problem of starvation. Some alternatives for the validation-phase are considered that not only avoid these disadvantages but also improve the performance of the original approach. Most promising is a technique which supports read-transactions, thereby decreasing the number of backups substantially. An adaptation of a centralized optimistic concurrency control scheme to a distributed environment is presented. The proposed algorithm is deadlock-free and offers high parallelism.< >
A parallel computation model is introduced that is suitable for pursuit of large-scale concurrency. The goal is to develop a semantically clean paradigm for distributed computation with fine-grained parallelism. The a...
详细信息
A parallel computation model is introduced that is suitable for pursuit of large-scale concurrency. The goal is to develop a semantically clean paradigm for distributed computation with fine-grained parallelism. The approach taken is to promote the notion of transaction, an anonymous atomic action void of internal state, as the basic element of computation. A computation is organized as a network, called a transaction net, of databases connected by transactions. A transaction, when it is fired, consumers data objects from source databases and produces data objects in target databases as an atomic action. A transaction net is akin to a Petri net, where the token, the place, and the transition corresponds to the data, the database, and the transaction, respectively. The state of computation is represented by the data state without the control state. An informal definition of the model is given. Solutions are given for well-known programming problems.< >
A scrambled storage scheme is proposed for storage arrays of N*N elements in N=2/sup n/ parallel memory-modules to allow conflict-free access to various array partitions. It is shown that the scheme allows conflict-fr...
详细信息
A scrambled storage scheme is proposed for storage arrays of N*N elements in N=2/sup n/ parallel memory-modules to allow conflict-free access to various array partitions. It is shown that the scheme allows conflict-free access to rows, columns, square blocks, and distributed blocks of stored arrays. An alternative way of achieving the desired accessibility would use P. Budnik and D.J. Kuck's (1971) nonuniform skewed storage; in this case, addressing hardware would require O(n*2/sup n/) exclusive-OR circuits. The proposed scheme has, however, the advantage of simplifying address generation; addressing hardware requires n exclusive-OR circuits only. Some of the important questions of scrambling/unscrambling data through a proposed interconnection network are discussed.< >
A scrambled storage scheme is proposed for storing arrays of NXN elements in N = 2n parallel memory modules to allow conflict-free access to various array partitions. It is shown that the scheme allows conflict-free a...
ISBN:
(纸本)9780818608612
A scrambled storage scheme is proposed for storing arrays of NXN elements in N = 2n parallel memory modules to allow conflict-free access to various array partitions. It is shown that the scheme allows conflict-free access to rows, columns, square blocks, and distributed blocks of stored arrays. An alternative way of achieving the desired accessibility would use Budnik and Kuck's nonuniform skewed storage [3]; in this case, addressing hardware would require Ο(nX2n) exclusive-or circuits. The proposed scheme has, however, the advantage of simplifying address generation; addressing hardware requires n exclusive-or circuits only. Some of the important questions of scrambling/unscrambling data through a proposed interconnection network are discussed.
The following topics are dealt with: access methods;distributed operating systems and databases;database design and implementation;performance evaluation;architectural support for database management;evaluating recurs...
详细信息
ISBN:
(纸本)0818607629
The following topics are dealt with: access methods;distributed operating systems and databases;database design and implementation;performance evaluation;architectural support for database management;evaluating recursive queries;file structures;parallel processing database systems;object-based systems;performance in distributedsystems;improving concurrency in distributedsystems;fault tolerance and correctness;knowledge representation;resiliency in distributedsystems;fault-tolerant storage systems;data modeling;historical databases;extending the relational model;CAD/CAM systems;and query processing. 85 papers are published in the present proceedings.
The author discusses various aspects of distributed control systems: the multiple allocation procedure protocol;security;migration strategies;memory requirements;the use of parallel processing;open architectures;the d...
详细信息
The author discusses various aspects of distributed control systems: the multiple allocation procedure protocol;security;migration strategies;memory requirements;the use of parallel processing;open architectures;the development of low-level interfaces;the role of relational databases;intelligent transmitters;microcomputers;real-time characteristics;the man-machine interface;and programmable controllers.
The following topics are dealt with: networks and distributed processing;artificial intelligence;databases;computer systems;computer applications;parallel processing;algorithms and data structures;fault tolerance and ...
详细信息
ISBN:
(纸本)0818607807
The following topics are dealt with: networks and distributed processing;artificial intelligence;databases;computer systems;computer applications;parallel processing;algorithms and data structures;fault tolerance and reliability;performance evaluation;computer graphics;image processing and pattern recognition;computer-aided engineering;office automation;design and test of VLSI;and software and tools. 154 papers were presented of which 140 are published in the present proceedings.
暂无评论