The author presents a practical sorting algorithm for a nearest-neighbor connected array of MIMD (multiple-instruction, multiple-data-stream) processors, each handling a significant fraction of the elements to be sort...
详细信息
ISBN:
(纸本)0818608722
The author presents a practical sorting algorithm for a nearest-neighbor connected array of MIMD (multiple-instruction, multiple-data-stream) processors, each handling a significant fraction of the elements to be sorted. The algorithm is communication and CPU-optimal for a linear array, is almost order-preserving, and requires little working memory.
A parallel algorithm for sorting is developed which has a time complexity of O(log n) and requires n**2/log n processors. The algorithm can be readily mapped on an SIMD (single-instruction, multiple-data-stream) mesh-...
详细信息
ISBN:
(纸本)0818608307
A parallel algorithm for sorting is developed which has a time complexity of O(log n) and requires n**2/log n processors. The algorithm can be readily mapped on an SIMD (single-instruction, multiple-data-stream) mesh-connected array of processors which has all the features of efficient VLSI implementation. The corresponding hardware algorithm maintains the O(log n) execution time and has a low, O(n) interprocessor communication time.
The author suggests two uses for user protocols: as litmus tests for document (and interface) decisions being made during development, and as agents for building a long-term understanding of users as actors. She also ...
详细信息
The author suggests two uses for user protocols: as litmus tests for document (and interface) decisions being made during development, and as agents for building a long-term understanding of users as actors. She also reviews the various history of usability studies and presents extent definitions of protocols.
The authors present a representation for concurrent systems, called a task interaction graph, that facilitates analysis. They have developed a model of interacting tasks that can considerably reduce the number of stat...
详细信息
ISBN:
(纸本)0818608684
The authors present a representation for concurrent systems, called a task interaction graph, that facilitates analysis. They have developed a model of interacting tasks that can considerably reduce the number of states in concurrency graph representations. They call this representation a task interaction concurrency graph (TIGC), since it is derived from a task interaction graph (TIG) instead of from a control-flow representation. Using their model, the authors have compared the resulting representations for some of the common concurrency examples that appear in the literature. For these examples, the number of states were reduced by well over 50%. Such a substantial reduction is expected to have a major impact on the kinds of analysis that can be applied and on the kinds of programs that can be analyzed.
The design implementation of a sequential circuit which compares and sorts eight words simultaneously is presented. Each word is an unsigned integer of K bits, where K can be any positive integer. The words are fed in...
详细信息
ISBN:
(纸本)0818608471
The design implementation of a sequential circuit which compares and sorts eight words simultaneously is presented. Each word is an unsigned integer of K bits, where K can be any positive integer. The words are fed into the sorter starting with the most significant bit of each word. On each clock pulse one bit from each word is received by the sorter and they are routed to their proper destination. The sorter is implemented using complementary metal-oxide-silicon (CMOS) technology.
Concurrent programs are analyzed by deriving constraints on the feasible SYN-sequences of a concurrent program according to the program's syntactic and semantic information. These constraints, called feasibility c...
详细信息
ISBN:
(纸本)0818608684
Concurrent programs are analyzed by deriving constraints on the feasible SYN-sequences of a concurrent program according to the program's syntactic and semantic information. These constraints, called feasibility constraints, show restrictions on the ordering of synchronization events allowed by the program. By using feasibility constraints, one can obtain a better approximation of the feasibility set of a concurrent program and improve the effectiveness of error detection by static analysis.
MMX (Multiprocessor Multitasking eXecutive) is a small yet powerful operating system for shared-memory multiprocessors. The MMX parallel processor is a small shared-bus multiprocessor assembled from several commercial...
详细信息
MMX (Multiprocessor Multitasking eXecutive) is a small yet powerful operating system for shared-memory multiprocessors. The MMX parallel processor is a small shared-bus multiprocessor assembled from several commercial processor boards. Together, MMX and its parallel processor provide a flexible and powerful testbed for parallel software development. A description is given of MMX design principles, its structure and services. Parallel programming techniques with MMX are demonstrated, with special emphasis on work distribution. The author concludes with some MMX timing and speedup measurements of several parallel programs.
A formal definition is presented of the semantics of concurrent systems with priority constraints. It is shown that neither partial orders, which form the basis of several formal semantical theories which explicitly m...
详细信息
A formal definition is presented of the semantics of concurrent systems with priority constraints. It is shown that neither partial orders, which form the basis of several formal semantical theories which explicitly model concurrency, nor total orders, which form the basis of all formal semantical theories which reduce concurrency to arbitrary interleaving, suffice for the treatment of priority. A dialect of COSY, a notation for specifying concurrent systems whose formal semantics is based on partial orders for the treatment of systems in the absence of priorities, is used. A semantical theory of COSY systems including priority constraints, which is based on sequences of multiples is introduced. Finally, it is shown how the concept of starvation can be formally defined in the approach presented.
A tree-connected processor system (TPS) is one where m (≥ 1) processors are connected to constitute a rooted tree. The authors study the problem of finding a feasible schedule for a TPS for n jobs with release times ...
详细信息
ISBN:
(纸本)9517212402
A tree-connected processor system (TPS) is one where m (≥ 1) processors are connected to constitute a rooted tree. The authors study the problem of finding a feasible schedule for a TPS for n jobs with release times and deadlines. It is shown that the problem can be solved in O(mn log n) time if jobs have unit processing times.
Even within the current drive for quality, documentation quality is often ignored or forgotten. This article cites a number of ways of improving the quality of documentation and, in particular, describes and discusses...
详细信息
Even within the current drive for quality, documentation quality is often ignored or forgotten. This article cites a number of ways of improving the quality of documentation and, in particular, describes and discusses the inspection method.
暂无评论