Search results for " computing"
showing 10 items of 2075 documents
An optimized path-selection using airtime metric in OLSR networks: Implementation and testing
2008
Wireless mesh networks used for broadband access may be equipped with multiple Internet gateways. Using airtime metric as the main criterion for path selection is an effective method for achieving load-balancing in such a multi-homed mesh network. In this paper, we describe how the radio-aware path selection using airtime as a metric can be implemented in order to achieve this functionality, based on a modified version of an open source OLSR implementation. The paper presents both the principle of this method, its implementation and the empirical results which focus on testing the airtime metric functionality based on a deployed small-scale test network.
Performance Analysis in Spatially Correlated IEEE 802.11 Networks
2012
Wireless mesh networks are difficult to be characterized, especially under multi-hop traffic streams. The problem is that the local view of the channel and the correlation between the buffers of consecutive nodes in a stream path make complicated the identification of the contention level perceived by each station along the time. Such a figure is used in the models based on the so called decoupling assumption for evaluating the final scheduling of simultaneous channel access grants. In this paper we propose a simplified mesh network model focused on capturing the correlation due to the network topology and traffic routes rather than the access protocol state at each node. To this purpose, w…
An adaptive timer for RPL to handle mobility in wireless sensor networks
2016
This paper focuses on the performance of wireless sensor networks characterized by a hybrid topology composed of mobile and static sensor nodes. The Routing Protocol for Low power and lossy networks (RPL), which is standardized as an IPv6 routing protocol for low power and lossy networks, uses the trickle timer algorithm to handle changes in the network topology. However, this algorithm is not well adapted to dynamic environments. This paper enhances the trickle timer in order to fit with mobility requirements. Most of previous works have improved this algorithm without considering the random movement of nodes. In this work, the proposed timer algorithm takes into consideration the random t…
Throughput and energy efficiency comparison of one-hop, two-hop, virtual relay and cooperative retransmission schemes
2010
Two main types of approaches exist for implementing cooperative communications at the MAC layer: virtual-hop relay and cooperative retransmission. While the virtual-hop relay schemes employ relay nodes to forward packets when higher end-to-end throughput can be achieved compared with the direct transmission, the cooperative retransmission schemes use relays to retransmit data only after the direct transmission fails. However, the performance of the these different approaches has not been compared in the literature, especially when energy efficiency is considered. In order to find out the best transmission scheme, this paper evaluates and compares the performance of the one-hop direct transm…
Latency factor in worldwide IP routed networks
2014
Current Border Gateway Protocol (BGP) serves for worldwide internet backbone connections between different Autonomous Systems (AS) on the routing level. Continuously in real-time changing environment core routers calculate routing decisions based on path-vector database entries. AS path-vector database may contain multiple paths for various Internet Protocol (IP) prefix destinations where smallest entity in the path-vector is AS. In real world AS can be internet service operator, university or any other organization willing to participate in internet network via multi-homed uplinks. Unfortunately, internal part of AS is hidden in path-vector database thus excluding factors like latency, pac…
Modeling Networks of Probabilistic Memristors in SPICE
2021
Efficient simulation of stochastic memristors and their networks requires novel modeling approaches. Utilizing a master equation to find occupation probabilities of network states is a recent major departure from typical memristor modeling [Chaos, solitons fractals 142, 110385 (2021)]. In the present article we show how to implement such master equations in SPICE – a general purpose circuit simulation program. In the case studies we simulate the dynamics of acdriven probabilistic binary and multi-state memristors, and dc-driven networks of probabilistic binary and multi-state memristors. Our SPICE results are in perfect agreement with known analytical solutions. Examples of LTspice code are…
Language-Integrated Privacy-Aware Distributed Queries
2019
Distributed query processing is an effective means for processing large amounts of data. To abstract from the technicalities of distributed systems, algorithms for operator placement automatically distribute sequential data queries over the available processing units. However, current algorithms for operator placement focus on performance and ignore privacy concerns that arise when handling sensitive data. We present a new methodology for privacy-aware operator placement that both prevents leakage of sensitive information and improves performance. Crucially, our approach is based on an information-flow type system for data queries to reason about the sensitivity of query subcomputations. Ou…
Examining business models of Software-as-a-Service companies
2013
Software-as-a-Service Revenue Models
2013
When should software providers maintain their traditional licensing model versus offering software as a service, and which SaaS model is more profitable: rental or pay per use? For customers, what are the trade-offs between traditional licensing, renting, and paying per use? peerReviewed
Scalable Dense Factorizations for Heterogeneous Computational Clusters
2008
This paper discusses the design and the implementation of the LU factorization routines included in the Heterogeneous ScaLAPACK library, which is built on top of ScaLAPACK. These routines are used in the factorization and solution of a dense system of linear equations. They are implemented using optimized PBLAS, BLACS and BLAS libraries for heterogeneous computational clusters. We present the details of the implementation as well as performance results on a heterogeneous computing cluster.