Generic low-latency NoC router architecture for FPGA computing systems

A novel cost-effective and low-latency wormhole router for packet-switched NoC designs, tailored for FPGA, is presented. This has been designed to be scalable at system level to fully exploit the characteristics and constraints of FPGA based systems, rather than custom ASIC technology. A key feature is that it achieves a low packet propagation latency of only two cycles per hop including both router pipeline delay and link traversal delay - a significant enhancement over existing FPGA designs - whilst being very competitive in terms of performance and hardware complexity. It can also be configured in various network topologies including 1-D, 2-D, and 3-D. Detailed design-space exploration has been carried for a range of scaling parameters, with the results of various design trade-offs being presented and discussed. By taking advantage of abundant buildin reconfigurable logic and routing resources, we have been able to create a new scalable on-chip FPGA based router that exhibits high dimensionality and connectivity. The architecture proposed can be easily migrated across many FPGA families to provide flexible, robust and cost-effective NoC solutions suitable for the implementation of high-performance FPGA computing systems. © 2011 IEEE.

Swarm scheduling approaches for work-flow applications with security constraints in distributed data-intensive computing environments

The scheduling problem in distributed data-intensive computing environments has become an active research topic due to the tremendous growth in grid and cloud computing environments. As an innovative distributed intelligent paradigm, swarm intelligence provides a novel approach to solving these potentially intractable problems. In this paper, we formulate the scheduling problem for work-flow applications with security constraints in distributed data-intensive computing environments and present a novel security constraint model. Several meta-heuristic adaptations to the particle swarm optimization algorithm are introduced to deal with the formulation of efficient schedules. A variable neighborhood particle swarm optimization algorithm is compared with a multi-start particle swarm optimization and multi-start genetic algorithm. Experimental results illustrate that population based meta-heuristics approaches usually provide a good balance between global exploration and local exploitation and their feasibility and effectiveness for scheduling work-flow applications. © 2010 Elsevier Inc. All rights reserved.

Effectiveness of granular columns in containing settlement

Laboratory-based research studies and full-scale evaluations of the behaviour of ground improved with granular columns are ample regarding bearing capacity, but limited in respect to the settlement response. This paper presents a laboratory model study that considers the settlement performance of isolated pad footings bearing on reinforced sand deposits under the influence of a fluctuating groundwater table. This is a particularly onerous condition for loose sand deposits in coastal areas which may undergo significant collapse settlement over time. Loose and dense experimental sand beds were constructed and the performance of rigid footings under a maintained load and bearing on sand incorporating different column configurations were monitored under cycling of the water table over a period of 28 days, with one filling/empting cycle every 18 h. It was found that settlement, while greatly reduced compared with unreinforced footings, was ongoing and typically occurred at a much greater rate for loose sand than dense sand. Also, settlement rates were slightly higher for fully penetrating than partially penetrating columns and also for footings reinforced by a column group rather than a single column. This was attributed to the migration of sand grains into the larger column voids.

A laboratory based model study. Experimental observations of footings supported on soft clay reinforced with granular columns.

The inclusion of granular columns in soft clay deposits leads to improvements in bearing capacity and overall stiffness along with a reduction in consolidation settlement. Many laboratory investigations have focused on aspects of bearing capacity, but published data on settlement performance is limited. This paper reports on some interesting findings obtained from a laboratory model study in respect of these issues. In this investigation, 300 mm diameter by 400 mm long samples of soft kaolin clay were reinforced with single or multiple granular columns of various lengths using the displacement and replacement installation methods. The experimental findings revealed that, for the same area replacement ratio, limited settlement reduction was achieved for single long floating columns and end-bearing column groups. Marginal improvements in settlement performance were also achieved for columns installed by the displacement method. No settlement reduction was achieved for short single floating columns while short floating granular column groups produced increased settlements. These observations were verified using contact pressure measurements between the footing and column/surrounding clay.