Search

In this paper, several efficient migration and allocation strategies have been compared on the mesh-based multiprocessor systems. The traditional non-preemptive submesh allocation strategies consist of two row boundary (TRB) and two column boundary (TCB). The existing migration mechanisms are online dynamic compaction-four corner (ODC-FC), limited top-down compaction (LTDC), TCB, and the combination of TCB and ODC-FC algorithms. Indeed, the new allocation method is presented in this paper. This mechanism has the benefits of two efficient traditional allocation algorithms. It is the combination of the TCB and TRB allocation methods. Also, in this process the impact of four key metrics on online mapping is considered. The parameters are average task execution time (ATET), average task system utilization (ATSU), average task waiting time (ATWT), and average task response time (ATRT). Using TCB and TRB mechanism with the migration strategies is shown that the new algorithm has better ATET, ATRT, ATWT, and ATSU. It has, respectively, 23.5494, 97.1216, 39.1291, and 4.142% improvements in comparison with the previous mechanisms. [ABSTRACT FROM AUTHOR]

We characterize effect algebras and pseudo effect algebras that can be represented as an interval of the lexicographic product of an antilattice unital po-group ( H, u) with a directed po-group G, both with the Riesz Decomposition Property (RDP). We show that a crucial condition is the existence of a lexicographic normal ideal. Finally, we present a categorical equivalence of the category of ( H, u)-lexicographic pseudo effect algebras having RDP with the category of directed po-groups with RDP. In addition, a weaker form of the ( H, u)-lexicographic pseudo effect algebra is studied. [ABSTRACT FROM AUTHOR]

The efficiency of the k-Nearest Neighbour classifier depends on the size of the training set as well as the level of noise in it. Large datasets with high level of noise lead to less accurate classifiers with high computational cost and storage requirements. The goal of editing is to improve accuracy by improving the quality of the training datasets. To obtain such datasets, editing removes noise and mislabeled data as well as smooths the decision boundaries between the discrete classes. On the other hand, prototype abstraction aims to reduce the computational cost and the storage requirements of classifiers by condensing the training data. This paper proposes an editing algorithm called Editing through Homogeneous Clusters (EHC). Then, it extends the idea by introducing a prototype abstraction algorithm that integrate the EHC mechanism and is capable of creating a small noise-free representative set of the initial training data. This algorithm is called Editing and Reduction through Homogeneous Clusters (ERHC). Both are based on a fast and parameter free iterative execution of k-means clustering that forms homogeneous clusters. Both consider as noise and remove clusters consisting of a single item. In addition, ERHC summarizes the items of the remaining clusters by storing the mean item for each one in the representative set. EHC and ERHC are tested on several datasets. The results show that both run very fast and achieve high accuracy. In addition, ERHC achieves high reduction rates. [ABSTRACT FROM AUTHOR]

We give an account on the authors' experience and results from the software verification competition held at the Formal Methods 2012 conference. Competitions like this are meant to provide a benchmark for verification systems. It consisted of three algorithms which the authors have implemented in Java, specified with the Java Modeling Language, and verified using the KeY system. Building on our solutions, we argue that verification systems which target implementations in real-world programming languages better have powerful abstraction capabilities. Regarding the KeY tool, we explain features which, driven by the competition, have been freshly implemented to accommodate for these demands. [ABSTRACT FROM AUTHOR]