When multiple central processing unit(CPU)cores and integrated graphics processing units(GPUs)share off-chip main memory,CPU and GPU applications compete for the critical memory resource.This causes serious resource c...When multiple central processing unit(CPU)cores and integrated graphics processing units(GPUs)share off-chip main memory,CPU and GPU applications compete for the critical memory resource.This causes serious resource competition and has a negative impact on the overall performance of the system.We describe the competition for shared-memory resources in a CPU-GPU heterogeneous multi-core architecture,and a sharedmemory request scheduling strategy based on perceptual and predictive batch-processing is proposed.By sensing the CPU and GPU memory request conditions in the request buffer,the proposed scheduling strategy estimates the GPU latency tolerance and reduces mutual interference between CPU and GPU by processing CPU or GPU memory requests in batches.According to the simulation results,the scheduling strategy improves CPU performance by8.53%and reduces mutual interference by 10.38%with low hardware complexity.展开更多
Combining different independent cloud services must coordinate their access control policies. Otherwise unauthorized access to composite cloud service can occur when there＇s a conflict among different cloud service p...Combining different independent cloud services must coordinate their access control policies. Otherwise unauthorized access to composite cloud service can occur when there＇s a conflict among different cloud service providers＇ access control policies, and then it will bring serious data security and privacy issues. In this paper, we propose Packet, a novel access control policy composition method that can detect and resolve policy conflicts in cloud service composition, including those conflicts related to privacyaware purposes and conditions. The Packet method is divided into four steps. First, employing a unified description, heterogeneous policies are transformed into a unified attributebased format. Second, to improve the conflict detection ef- ficiency, policy conflicts on the same resource can be eliminated by adopting cosine similarity-based algorithm. Third, exploiting a hierarchical structure approach, policy conflicts related to different resources or privacy-aware purposes and conditions can be detected. Fourth, different conflict resolution techniques are presented based on the corresponding conflict types. We have successfully implemented the Packet method in Openstack platform. Comprehensive experiments have been conducted, which demonstrate the effectiveness of the proposed method by the comparison with the existing XACML-based system at conflict detection and resolution performance.展开更多
大数据时代,需要集成多样化数据管理分析工具完成业务需求,然而不同工具的数据访问接口标准不一,需要借助适配器进行接口转换以实现数据管理分析工具协作.因此,如何快速集成不同工具成为学术界与工业界亟待解决的问题.本文提出了一种基...大数据时代,需要集成多样化数据管理分析工具完成业务需求,然而不同工具的数据访问接口标准不一,需要借助适配器进行接口转换以实现数据管理分析工具协作.因此,如何快速集成不同工具成为学术界与工业界亟待解决的问题.本文提出了一种基于查询语言转换的多源数据统一访问框架BAF4DUA(Bi-Adapter Framework for Data Unified Access),该框架采用双端接口适配方式,在数据提供者工具端和数据消费者工具端分别引入适配器对系统数据与查询语义进行适配,将查询语言与数据存储模型相分离,从而实现了数据提供者与数据消费者之间多对多、即插即用的数据访问,提高了应用的灵活性与扩展性.展开更多
基金Project supported by the National Natural Science Foundation of China(Nos.62276011 and 61202076)the Natural Science Foundation of Beijing,China(No.4192007)。
文摘When multiple central processing unit(CPU)cores and integrated graphics processing units(GPUs)share off-chip main memory,CPU and GPU applications compete for the critical memory resource.This causes serious resource competition and has a negative impact on the overall performance of the system.We describe the competition for shared-memory resources in a CPU-GPU heterogeneous multi-core architecture,and a sharedmemory request scheduling strategy based on perceptual and predictive batch-processing is proposed.By sensing the CPU and GPU memory request conditions in the request buffer,the proposed scheduling strategy estimates the GPU latency tolerance and reduces mutual interference between CPU and GPU by processing CPU or GPU memory requests in batches.According to the simulation results,the scheduling strategy improves CPU performance by8.53%and reduces mutual interference by 10.38%with low hardware complexity.
文摘Combining different independent cloud services must coordinate their access control policies. Otherwise unauthorized access to composite cloud service can occur when there＇s a conflict among different cloud service providers＇ access control policies, and then it will bring serious data security and privacy issues. In this paper, we propose Packet, a novel access control policy composition method that can detect and resolve policy conflicts in cloud service composition, including those conflicts related to privacyaware purposes and conditions. The Packet method is divided into four steps. First, employing a unified description, heterogeneous policies are transformed into a unified attributebased format. Second, to improve the conflict detection ef- ficiency, policy conflicts on the same resource can be eliminated by adopting cosine similarity-based algorithm. Third, exploiting a hierarchical structure approach, policy conflicts related to different resources or privacy-aware purposes and conditions can be detected. Fourth, different conflict resolution techniques are presented based on the corresponding conflict types. We have successfully implemented the Packet method in Openstack platform. Comprehensive experiments have been conducted, which demonstrate the effectiveness of the proposed method by the comparison with the existing XACML-based system at conflict detection and resolution performance.
文摘大数据时代,需要集成多样化数据管理分析工具完成业务需求,然而不同工具的数据访问接口标准不一,需要借助适配器进行接口转换以实现数据管理分析工具协作.因此,如何快速集成不同工具成为学术界与工业界亟待解决的问题.本文提出了一种基于查询语言转换的多源数据统一访问框架BAF4DUA(Bi-Adapter Framework for Data Unified Access),该框架采用双端接口适配方式,在数据提供者工具端和数据消费者工具端分别引入适配器对系统数据与查询语义进行适配,将查询语言与数据存储模型相分离,从而实现了数据提供者与数据消费者之间多对多、即插即用的数据访问,提高了应用的灵活性与扩展性.
