Large-scale graph structures are considered as a keystone for many emerging high-performance computing applications in which Breadth-First Search (BFS) is an important building block. For such graph structures, BFS operations tends to be memory-bound rather than compute-bound. In this paper, we present an efficient reconfigurable architecture for parallel BFS that adopts new optimizations for utilizing memory bandwidth. Our architecture adopts a custom graph representation based on compressed-sparse raw format (CSR), as well as a restructuring of the conventional BFS algorithm. By taking maximum advantage of available memory bandwidth, our architecture continuously keeps our processing elements active. Using a commercial high-performance reconfigurable computing system (the Convey HC-2), our results demonstrate a 5× speedup over previously published FPGA-based implementations.

Abstract-In this paper, we analyze the performance of cooperative content caching in vehicular ad hoc networks (VANETs). In particular, we characterize, using analysis and simulations, the behavior of the probability of outage (i.e. not finding a requested data chunk at a neighbor) under freeway vehicular mobility. First, we introduce a formal definition for the probability of outage in the context of cooperative content caching. Second, we characterize, analytically, the outage probability under vehicular and random mobility scenarios. Next, we verify the analytical results using simulations and compare the performance under a number of plausible mobility scenarios. This provides key insights into the problem and the involved trade-offs and enable us to assess the potential opportunity offered by the, somewhat structured, vehicular mobility that can be exploited by cooperative content caching schemes. The presented numerical results exhibit complete agreement between the analytical and simulation studies. Finally, we observe that vehicular mobility creates opportunities for enhanced outage performance under practically relevant scenarios.

No abstract

No abstract

No abstract