The paper deals with a method of decomposition of convex programming problems. This method admits an arbitrary split of the original problem into subproblems and may be applied to decentralized optimization of economi...
详细信息
The paper deals with a method of decomposition of convex programming problems. This method admits an arbitrary split of the original problem into subproblems and may be applied to decentralized optimization of economic and technical systems;it may be realized with the aid of several parallel computers. The method converges for almost arbitrary solvable convex programming problems. The idea of the method is based on a general decompositional approach to solving equations determined by a sum of several monotone mappings;this approach is described in Section 1 of the present paper.
Iterative algorithms are described which generate permutative schedules in information and computing systems of various configurations and networks. The algorithms' complexity is evaluated. They feature limited me...
详细信息
Iterative algorithms are described which generate permutative schedules in information and computing systems of various configurations and networks. The algorithms' complexity is evaluated. They feature limited memory requirements, a simple design of system models, and simple optimization procedures with various criteria.
There are many papers which are concerned with modeling a unibus homogeneous multiprocess system. For this kind of system, the machine servicing model has been used, and performance indices had been calculated (Kobaya...
详细信息
There are many papers which are concerned with modeling a unibus homogeneous multiprocess system. For this kind of system, the machine servicing model has been used, and performance indices had been calculated (Kobayashi 1978 and Marson 1981). For heterogeneous multiprocessor system, the BCMP class of queueing networks can be solved (F. Baskett et al. 1975). But it is not easy to calculate some performance indices. This paper uses approximate algorithm to estimate the mean number of effective processor in system, it is easy to calculate and can reach appropriate accuracy.
The increasing complexity of automation has led to greater demands for user-friendly documentation. Software documentation as it applies to the explanation, representation and combination of STEP 5 user programs is co...
详细信息
The increasing complexity of automation has led to greater demands for user-friendly documentation. Software documentation as it applies to the explanation, representation and combination of STEP 5 user programs is considered.
The model we consider is the (concurrent-write, PRIORITY) PRAM. It has n synchronous processors, which communicate via an infinite shared memory. When several processors simultaneously write to the same cell, the one ...
详细信息
The model we consider is the (concurrent-write, PRIORITY) PRAM. It has n synchronous processors, which communicate via an infinite shared memory. When several processors simultaneously write to the same cell, the one with the largest index succeeds. We allow the processors arbitrary computational power. Our main result is that sorting n integers requires OMEGA ( ROOT log n) steps in this strong model. This bound is proved in two stages. First, using a novel Ramsey theoretic argument, we 'reduce' sorting on a PRAM to sorting on a parallel merge tree. This tree is a generalization of Valiant's parallel comparison tree from left bracket V right bracket in which at every step n pairs of (previously ordered) sets are merged (rather then n pairs of elements compared). The second stage is proving the lower bound for such trees.
We adapt several parallel sorting algorithms (block sorting algorithms) and distributed sorting algorithms for implementation on an Ethermet network with diskless Sun workstations. We argue that the performance of sor...
详细信息
ISBN:
(纸本)089791225X
We adapt several parallel sorting algorithms (block sorting algorithms) and distributed sorting algorithms for implementation on an Ethermet network with diskless Sun workstations. We argue that the performance of sorting algorithms on local area networks (LANs) should be analyzed in a manner that is different from the ways that parallel and distributed sorting algorithms are usually analyzed. Consequently, we propose an empirical approach which will provide more insight into the performance of the algorithms. We obtain data on communication time, local processing time, and response time (i. e. total running time) of each algorithm for various file sizes and different numbers of processors. Comparing the performance data with our theoretical analysis, we attempt to provide rationale for the behaviour of the algorithms and project the future behaviour of the algorithms as file size, number of processors, or interprocessor communication facilities change.
We discuss time lower bounds for iterative merge sorts and iterative pseudo-merge sorts on an n multiplied by n mesh-connected processor array. They are 4. 5n-3log//2n minus 2 steps and 35n-log//2n-3 steps for sorting...
详细信息
We discuss time lower bounds for iterative merge sorts and iterative pseudo-merge sorts on an n multiplied by n mesh-connected processor array. They are 4. 5n-3log//2n minus 2 steps and 35n-log//2n-3 steps for sorting n**2 items. We also discuss time lower bounds for these sorting algorithms on higher dimensional models. For the case of the three-dimensional mesh-connected model, 7. 25n-4log//2n minus 8 steps and 5. 25n minus log//2n minus 6 steps are lower bounds for sorting n**3 items by iterative merge sorts and iterative pseudo-merge sorts, respectively.
A method is suggested for evaluating the quality of multiprogramming job mixes with the use of homogeneous and inhomogeneous network models of a computer system (CC). The results of an application of the models to an ...
详细信息
A method is suggested for evaluating the quality of multiprogramming job mixes with the use of homogeneous and inhomogeneous network models of a computer system (CC). The results of an application of the models to an actual CC are reported.
A randomized parallel algorithm that sorts on an N node network with constant valence in O(log N) time is given. More particularly, the algorithm sorts N items on an N-node cube-connected cycles graph, and, for some c...
详细信息
A randomized parallel algorithm that sorts on an N node network with constant valence in O(log N) time is given. More particularly, the algorithm sorts N items on an N-node cube-connected cycles graph, and, for some constant k, for all large enough alpha , it terminates with k alpha log N time with probability at least 1 minus N** minus ** alpha .
Expressions are presented for RPS reconnect delays in three basic cases: single path, multiple path with static reconnect, multiple path with dynamic reconnect. The assumption of homogeneous reconnects, which is intro...
详细信息
ISBN:
(纸本)089791225X
Expressions are presented for RPS reconnect delays in three basic cases: single path, multiple path with static reconnect, multiple path with dynamic reconnect. The assumption of homogeneous reconnects, which is introduced in the analysis, is shown to be implicit in many prior analyses. This assumption simplifies the resulting equations, but more general equations are also presented for the case where homogeneous reconnects are not assumed. These general results have not appeared previously. This paper also uses the assumption of constrained independence to derive a result for static reconnect which has only been derived previously using the maximum entropy principle. In the case of dynamic reconnect, constrained independence yields an entirely new closed form result.
暂无评论