为提升大规模多维数据集的skyline计算效率,提出了一种多核并行算法MPSCS(multi-core parallelskyline computation based on sorting).首先按照任意一维对数据集进行预排序,然后划分为多个子集,使用skeleton并行程序设计模型进行并行...为提升大规模多维数据集的skyline计算效率,提出了一种多核并行算法MPSCS(multi-core parallelskyline computation based on sorting).首先按照任意一维对数据集进行预排序,然后划分为多个子集,使用skeleton并行程序设计模型进行并行化处理.与未采用预排序策略的多核并行算法相比,MPSCS算法处理过程简单,具有较好的渐进性、用户友好性和效率.实验结果表明,对规模较大、维数较高的数据集,效率可提高30%~40%,相对加速比可达线性.展开更多
HGGF(halo-based galaxy group finder)算法实现了基于暗物质晕的星系找群,在研究宇宙大尺度结构及宇宙的演化等领域中占有至关重要的地位。但由于数据规模的增长,急需对HGGF算法进行优化,以缩短运行时间。经分析,算法的热点部分耗时受...HGGF(halo-based galaxy group finder)算法实现了基于暗物质晕的星系找群,在研究宇宙大尺度结构及宇宙的演化等领域中占有至关重要的地位。但由于数据规模的增长,急需对HGGF算法进行优化,以缩短运行时间。经分析,算法的热点部分耗时受到非规则访存的严重影响,因此针对算法的结构和非规则访存模型,提出了数据预排序方法,并分析了该方法如何影响访存过程。在此基础上,利用数据对齐、循环分解进一步优化访存效率,利用负载均衡和互斥变量私有化的方法提高了Open MP的并行效率,最终将HGGF应用使用12线程加速11.6倍,同时取得了更好的可扩展性。主要有三点贡献:(1)分析了HGGF算法的非规则访存问题;(2)提出并分析了数据预排序方法;(3)使用数据对齐、循环分解、负载均衡、互斥变量私有化方法提高了HGGF应用的并行性能。展开更多
Destination prediction has attracted widespread attention because it can help vehicle-aid systems recommend related services in advance to improve user driving experience.However,the relevant research is mainly based ...Destination prediction has attracted widespread attention because it can help vehicle-aid systems recommend related services in advance to improve user driving experience.However,the relevant research is mainly based on driving trajectory of vehicles to predict the destinations,which is challenging to achieve the early destination prediction.To this end,we propose a model of early destination prediction,DP-BPR,to predict the destinations by users’travel time and locations.There are three challenges to accomplish the model:1)the extremely sparse historical data make it challenge to predict destinations directly from raw historical data;2)the destinations are related to not only departure points but also departure time so that both of them should be taken into consideration in prediction;3)how to learn destination preferences from historical data.To deal with these challenges,we map sparse high-dimensional data to a dense low-dimensional space through embedding learning using deep neural networks.We learn the embeddings not only for users but also for locations and time under the supervision of historical data,and then use Bayesian personalized ranking(BPR)to learn to rank destinations.Experimental results on the Zebra dataset show the effectiveness of DP-BPR.展开更多
文摘为提升大规模多维数据集的skyline计算效率,提出了一种多核并行算法MPSCS(multi-core parallelskyline computation based on sorting).首先按照任意一维对数据集进行预排序,然后划分为多个子集,使用skeleton并行程序设计模型进行并行化处理.与未采用预排序策略的多核并行算法相比,MPSCS算法处理过程简单,具有较好的渐进性、用户友好性和效率.实验结果表明,对规模较大、维数较高的数据集,效率可提高30%~40%,相对加速比可达线性.
文摘HGGF(halo-based galaxy group finder)算法实现了基于暗物质晕的星系找群,在研究宇宙大尺度结构及宇宙的演化等领域中占有至关重要的地位。但由于数据规模的增长,急需对HGGF算法进行优化,以缩短运行时间。经分析,算法的热点部分耗时受到非规则访存的严重影响,因此针对算法的结构和非规则访存模型,提出了数据预排序方法,并分析了该方法如何影响访存过程。在此基础上,利用数据对齐、循环分解进一步优化访存效率,利用负载均衡和互斥变量私有化的方法提高了Open MP的并行效率,最终将HGGF应用使用12线程加速11.6倍,同时取得了更好的可扩展性。主要有三点贡献:(1)分析了HGGF算法的非规则访存问题;(2)提出并分析了数据预排序方法;(3)使用数据对齐、循环分解、负载均衡、互斥变量私有化方法提高了HGGF应用的并行性能。
基金Project(2018YFF0214706)supported by the National Key Research and Development Program of ChinaProject(cstc2020jcyj-msxmX0690)supported by the Natural Science Foundation of Chongqing,China+1 种基金Project(2020CDJ-LHZZ-039)supported by the Fundamental Research Funds for the Central Universities of Chongqing,ChinaProject(cstc2019jscx-fxydX0012)supported by the Key Research Program of Chongqing Technology Innovation and Application Development,China。
文摘Destination prediction has attracted widespread attention because it can help vehicle-aid systems recommend related services in advance to improve user driving experience.However,the relevant research is mainly based on driving trajectory of vehicles to predict the destinations,which is challenging to achieve the early destination prediction.To this end,we propose a model of early destination prediction,DP-BPR,to predict the destinations by users’travel time and locations.There are three challenges to accomplish the model:1)the extremely sparse historical data make it challenge to predict destinations directly from raw historical data;2)the destinations are related to not only departure points but also departure time so that both of them should be taken into consideration in prediction;3)how to learn destination preferences from historical data.To deal with these challenges,we map sparse high-dimensional data to a dense low-dimensional space through embedding learning using deep neural networks.We learn the embeddings not only for users but also for locations and time under the supervision of historical data,and then use Bayesian personalized ranking(BPR)to learn to rank destinations.Experimental results on the Zebra dataset show the effectiveness of DP-BPR.