The UK e-Science EPSRC pilot project Meeting the Design Challenges of nanoCMOS Electronics (***) was funded to address the challenges facing the global electronics semiconductor industry caused by the decreasing size ...
详细信息
ISBN:
(纸本)9780889868205
The UK e-Science EPSRC pilot project Meeting the Design Challenges of nanoCMOS Electronics (***) was funded to address the challenges facing the global electronics semiconductor industry caused by the decreasing size of Complementary Metal Oxide Semiconductor (CMOS) transistors and the atomic variability present in devices manifest at these dimensions. Fundamental problems to be addressed include the modelling, understanding and predicting the effect of differences in the atomic structure of devices on their behaviour, and then using this information to guide electronic circuit and system designers who utilise CMOS components. In this paper we describe the e-Infrastructure that has been developed as part of the nanoCMOS project and outline how it supports large scale high performance computing (HPC) simulations of ensembles of devices which can subsequently be used to model and understand the impact that they have on very large electronic circuits. Key features of this e-Infrastructure include support for very large scale HPC utilization;dealing with federated data sets and associated metadata from multi-level simulations, and addressing challenges related to security and intellectual property protection of data, simulation codes and electronic designs as a whole.
Overlay multicast protocol constructs a virtual mesh spanning all member nodes of a multicast group and employs standard unicast routing to fulfill multicast functionality on application layer. The advantages of this ...
详细信息
ISBN:
(纸本)9780889866386
Overlay multicast protocol constructs a virtual mesh spanning all member nodes of a multicast group and employs standard unicast routing to fulfill multicast functionality on application layer. The advantages of this approach are simplicity and flexibility. However, efficiency and stability are the issues that must be addressed as the size of the multicast group grows in the mobile ad hoc networks (MANETs). In this paper, we propose an effective structure for overlay multicast to solve these problems in MANETs. Instead of using a spanning tree on the virtual mesh, we adopt a simple structure called MCore for multicast. An MCore is a path that minimizes the sum of the distances of all vertices to the path plus the length of the path. The MCore is more stable and easier to maintain than the spanning tree in MANETs. The simulation results show that our approach handles the flexibility and mobility issues in overlay multicast protocols effectively for large multicast group size.
In this paper, we consider flows requiring quantitative end-to-end QoS guarantees. We focus more particularly on the end-to-end response times of flows and their probability of meeting their end-to-end deadlines. We a...
详细信息
ISBN:
(纸本)9780889866379
In this paper, we consider flows requiring quantitative end-to-end QoS guarantees. We focus more particularly on the end-to-end response times of flows and their probability of meeting their end-to-end deadlines. We assume that packets are scheduled according to their fixed priority, reflecting the importance degree of the flow. If on a node two packets have the same fixed priority, the packet with the smallest relative deadline on the node considered is scheduled first. Therefore, this non-preemptive scheduling, called FP/DM, takes into account the deadline constraint. The deterministic approach, based on a worst case analysis, may lead to a bound on the flow end-to-end response times that can be reached infrequently. A network dimensioning based on this bound can be expensive in terms of resources. That is why we are interested in probabilistic QoS guarantees. We then evaluate the benefits brought by FP/DM with regard to FP/FIFO. With the deterministic approach, we compare the worst case response times and the laxities of the flows considered. With the probabilistic approach, we compare the probabilities of meeting the deadlines for the flows considered as well as the p-schedulabibility of the flow set.
Overlapping of communication and computation has been intuitively used for a long time to improve the performance of parallelcomputing. One of approaches which allow increasing the ability of this overlapping is to u...
详细信息
ISBN:
(纸本)0889865701
Overlapping of communication and computation has been intuitively used for a long time to improve the performance of parallelcomputing. One of approaches which allow increasing the ability of this overlapping is to use active message. This paper presents the study of the overlapping communication/computation performance of INUKTITUT, an interface based on active message. This interface allows effectively and easily carrying of parallel programming high-level environments such as Athapascan on the computing clusters. We will present in this paper different performance evaluations of INUKTITUT on overlapping potential.
This paper describes the idea of a multi-dimensional bucket index designed for efficient indexing of telemetric readings. This data structure answers spatio-temporal range queries concerning utility usage within user ...
详细信息
ISBN:
(纸本)088986568X
This paper describes the idea of a multi-dimensional bucket index designed for efficient indexing of telemetric readings. This data structure answers spatio-temporal range queries concerning utility usage within user selected region and time. It has a capability of adjusting to incoming data and therefore is suitable to process data of highly dynamic nature. A stochastic prediction method is used to estimate utility usage in the near future. The real world implementation employs mobile agent system network which facilitates effective workload balancing of network hosts.
Genetic algorithrms have been applied to solve the 2-page drawing problem successfully, but they work with one global population, so the search time and space are limited. parallelization provides an attractive prospe...
详细信息
ISBN:
(纸本)088986568X
Genetic algorithrms have been applied to solve the 2-page drawing problem successfully, but they work with one global population, so the search time and space are limited. parallelization provides an attractive prospect in improving the efficiency and solution quality of genetic algorithms. One of the most popular tools for parallelcomputing is Message Passing Interface (AIPI). In this paper, we present four island models of parallel Genetic Algorithms with MPI: island models with linear, grid, random graph topologies and island model with periodical synchronisation. We compare their efficiency and quality of solutions for the 2-page drawing problem on a variety of graphs.
Dynamic server resource allocation to services on networks, or utility computing, is a powerful technology to provide required computer resources for multiple service providers at low cost. Virtual machine (VM) techno...
详细信息
ISBN:
(纸本)9780889866386
Dynamic server resource allocation to services on networks, or utility computing, is a powerful technology to provide required computer resources for multiple service providers at low cost. Virtual machine (VM) technology can be combined with utility computing to further improve server resource utilization. An important technical issue about VM-based utility computing is optimal placement of VMs on physical server nodes, because performance of services may be seriously affected by VM placement. We address this issue by introducing an on-line placement algorithm on the basis of performance influence among services. We have implemented this algorithm and evaluated on a simulated environment. The results have shown that the proposed mechanism can get roughly 25% better score calculated by the placement rules compared with a random placement algorithm.
The severe energy constraints of wireless sensor networks (WSNs) require energy-efficient communication protocols in order to fulfill the objectives of the application. Cross-layer design is a technique which can pote...
详细信息
ISBN:
(纸本)9780889867048
The severe energy constraints of wireless sensor networks (WSNs) require energy-efficient communication protocols in order to fulfill the objectives of the application. Cross-layer design is a technique which can potentially be used to improve the overall performance of WSNs by way of jointly optimizing and exploiting the interactions between various layers of the network protocol stack. In this paper, we propose a cross-layer framework design for the Embedded Middleware in Mobility Applications (EMMA) project. This optimization agent based framework design provides efficient data exchange between the various protocols layers via a state repository to improve the performance of WSN applications in terms of memory consumption and processing overhead.
An efficient range query processing support is required for distributed Hash Table (DHT)-based P2P networks as consistent hashing destroys the inherent order of numeric keys. In this paper, we present a lightweight an...
详细信息
ISBN:
(纸本)9780889866386
An efficient range query processing support is required for distributed Hash Table (DHT)-based P2P networks as consistent hashing destroys the inherent order of numeric keys. In this paper, we present a lightweight and efficient mechanism, called Range Hash Tree (RHT), to support range queries on DHTs. In RHT, key space is partitioned into small ranges such that data items within one range can be stored on one node. To quickly resolve a range query, we introduce a scalable encoding algorithm that can represent the key space partitioning status with a small RHT Signature. Compared to other approaches, RHT provides a bounded query delay, which is independent of the size of range query and the number of matching data items. Our experiments show that RHT works efficiently on both uniform and much skewed data distributions.
暂无评论