Cache-aware Approach Research Articles

Dynamic Source Rate Control for Cache-Based Transport Protocol in Wireless Sensor Networks

Reliability of data transport in low power and lossy networks like Wireless Sensor Networks (WSN) is vital due to its constraint characteristics and high probability of packet loss from wireless environment and congestion. Data caching and transmission rate control mechanisms are independent ways of improving the reliability of transport protocols in WSN as they immediately respond to packet losses in the network. However, these two mechanisms are traditionally designed independently and an appropriate rate control that can potentially improve the performance of cache-based transport protocols has not yet been investigated. In this work, a dynamic source rate control algorithm that utilizes a cache-aware approach was developed wherein cache management policies are employed to notify the relative degree of packet losses and limit the transmission window size. A baseline cache-based transport protocol called DTSN+ is used to implement the rate control algorithm and evaluated under network scenarios that experience packet losses from contention and congestion. The dynamic cache-aware algorithm outperformed other transport protocols in terms of cache utilization by 30%. The algorithm also obtained outstanding throughput, transmission time and fairness performance as compared with the fixed DTSN+ and DTC protocols. This demonstrated that a cache-aware approach can improve the performance of a transport protocol during high levels of packet losses in the network. In the future, these results can serve as an underlying support in designing a new cache-aware congestion control framework that is suitable for transport protocols that employ the use of intermediate caching in WSN.

Fast Ranking with Additive Ensembles of Oblivious and Non-Oblivious Regression Trees

Learning-to-Rank models based on additive ensembles of regression trees have been proven to be very effective for scoring query results returned by large-scale Web search engines. Unfortunately, the computational cost of scoring thousands of candidate documents by traversing large ensembles of trees is high. Thus, several works have investigated solutions aimed at improving the efficiency of document scoring by exploiting advanced features of modern CPUs and memory hierarchies. In this article, we present Q uick S corer , a new algorithm that adopts a novel cache-efficient representation of a given tree ensemble, performs an interleaved traversal by means of fast bitwise operations, and supports ensembles of oblivious trees. An extensive and detailed test assessment is conducted on two standard Learning-to-Rank datasets and on a novel very large dataset we made publicly available for conducting significant efficiency tests. The experiments show unprecedented speedups over the best state-of-the-art baselines ranging from 1.9 × to 6.6 × . The analysis of low-level profiling traces shows that Q uick S corer efficiency is due to its cache-aware approach in terms of both data layout and access patterns and to a control flow that entails very low branch mis-prediction rates.