在当前数据量爆炸式增长的推动下,整个信息社会步入了以存储和计算为核心的智能化时代。以硅基晶体管器件为基础的传统存储与计算技术由于当前晶体管尺寸微缩受限,而难以满足目前高能效、智能化任务的需求。人们迫切需要去开发一些新原...在当前数据量爆炸式增长的推动下,整个信息社会步入了以存储和计算为核心的智能化时代。以硅基晶体管器件为基础的传统存储与计算技术由于当前晶体管尺寸微缩受限,而难以满足目前高能效、智能化任务的需求。人们迫切需要去开发一些新原理、新结构的电子器件以及新型的计算架构去实现更先进的存储与计算。具有电致电阻转变效应的阻变存储器(Resistive Random Access Memory,RRAM),又称忆阻器,由于具有结构简单、功耗低、速度快、易于三维集成和存算一体等特性,被认为是后摩尔时代中实现新型存储与计算的关键器件之一,也是目前国际前沿研究热点。展开更多
Recently the content centric networks(CCNs) have been advocated as a new solution to design future networks. In the CCNs, content and its interest are delivered over the content store and pending interest table, respe...Recently the content centric networks(CCNs) have been advocated as a new solution to design future networks. In the CCNs, content and its interest are delivered over the content store and pending interest table, respectively, where both have limited capacities. Therefore, how to design the corresponding algorithms to efficiently deliver content and inertest over them becomes an important issue. In this paper, based on the analysis of content distribution, status of content store, and pending interest, we propose a novel caching algorithm with which the resources of content store and pending interest table can be efficiently used. Simulation results prove that the proposal can outperform the conventional methods.展开更多
文摘在当前数据量爆炸式增长的推动下,整个信息社会步入了以存储和计算为核心的智能化时代。以硅基晶体管器件为基础的传统存储与计算技术由于当前晶体管尺寸微缩受限,而难以满足目前高能效、智能化任务的需求。人们迫切需要去开发一些新原理、新结构的电子器件以及新型的计算架构去实现更先进的存储与计算。具有电致电阻转变效应的阻变存储器(Resistive Random Access Memory,RRAM),又称忆阻器,由于具有结构简单、功耗低、速度快、易于三维集成和存算一体等特性,被认为是后摩尔时代中实现新型存储与计算的关键器件之一,也是目前国际前沿研究热点。
基金supported in part by the fundamental key research project of Shanghai Municipal Science and Technology Commission under grant 12JC1404201the Ministry of Education Research Fund-China Mobile(2012) MCM20121032
文摘Recently the content centric networks(CCNs) have been advocated as a new solution to design future networks. In the CCNs, content and its interest are delivered over the content store and pending interest table, respectively, where both have limited capacities. Therefore, how to design the corresponding algorithms to efficiently deliver content and inertest over them becomes an important issue. In this paper, based on the analysis of content distribution, status of content store, and pending interest, we propose a novel caching algorithm with which the resources of content store and pending interest table can be efficiently used. Simulation results prove that the proposal can outperform the conventional methods.
