In this paper,we propose a novel spatial data index based on Hadoop:HQ-Tree.In HQ-Tree,we use PR QuadTrec to solve the problem of poor efficiency in parallel processing,which is caused by data insertion order and spac...In this paper,we propose a novel spatial data index based on Hadoop:HQ-Tree.In HQ-Tree,we use PR QuadTrec to solve the problem of poor efficiency in parallel processing,which is caused by data insertion order and space overlapping.For the problem that HDFS cannot support random write,we propose an updating mechanism,called "Copy Write",to support the index update.Additionally,HQ-Tree employs a two-level index caching mechanism to reduce the cost of network transferring and I/O operations.Finally,we develop MapReduce-based algorithms,which are able to significantly enhance the efficiency of index creation and query.Experimental results demonstrate the effectiveness of our methods.展开更多
In-network caching and Interest packets aggregation are two important features of Content-Centric Networking(CCN).CCN routers can directly respond to the Interest request by Content Store(CS)and aggregate the same Int...In-network caching and Interest packets aggregation are two important features of Content-Centric Networking(CCN).CCN routers can directly respond to the Interest request by Content Store(CS)and aggregate the same Interest packets by Pending Interest Table(PIT).In this way,most popular content requests will not reach the origin content server.Thus,content providers will be unaware of the actual usages of their contents in network.This new network paradigm presents content providers with unprecedented challenge.It will bring a great impact on existing mature business model of content providers,such as advertising revenue model based on hits amount.To leverage the advantages of CCN and the realistic business needs of content providers,we explore the hits-based content provisioning mechanism in CCN.The proposed approaches can avoid the unprecedented impact on content providers' existing business model and promote content providers to embrace the real deployment of CCN network.展开更多
Due to large size and different popularity for different part of the video, most proxy caches for streaming medias cache only a part of the video. Thus, an accurate understanding on the internal popularity distributio...Due to large size and different popularity for different part of the video, most proxy caches for streaming medias cache only a part of the video. Thus, an accurate understanding on the internal popularity distribution of media objects in streaming applications is very important for the development of efficient cache mechanisms. This letter shows that the internal popularity of popular streaming media obeys a k-transformed Zipf-like distribution through analyzing two 6-month long traces recorded at different streaming video servers of an entertainment video-on-demand provider. This empirical model can be used to design an efficient cach- ing algorithm.展开更多
Content-Centric Networking is a novel future network architecture that attracts increasing research interests in recent years. In-network caching has been regarded as a prominent feature of Content-Centric Networking ...Content-Centric Networking is a novel future network architecture that attracts increasing research interests in recent years. In-network caching has been regarded as a prominent feature of Content-Centric Networking since it is able to reduce the network traffic, alleviate the server bottleneck and decrease the user access latency. However, the CCN default caching scheme results in a high caching redundancy, causing an urgent need for an efficient caching scheme. To address this issue, we propose a novel implicit cooperative caching scheme to efficiently reduce the caching redundancy and improve the cache resources utilization. The simulation results show that our design achieves a higher hit ratio and a shorter cache hit distance in comparison with the other typical caching schemes.展开更多
To reduce network redundancy,innetwork caching is considered in many future Internet architectures,such as Information Centric Networking.In in-network caching system,the item sojourn time of LRU(Least Recently Used) ...To reduce network redundancy,innetwork caching is considered in many future Internet architectures,such as Information Centric Networking.In in-network caching system,the item sojourn time of LRU(Least Recently Used) replacement policy is an important issue for two reasons:firstly,LRU is one of the most common used cache policy;secondly,item sojourn time is positively correlated to the hit probability,so this metric parameter could be useful to design the caching system.However,to the best of our knowledge,the sojourn time hasn't been studied theoretically so far.In this paper,we first model the LRU cache policy by Markov chain.Then an approximate closedform expression of the item expectation sojourn time is provided through the theory of stochastic service system,which is a function of the item request rates and cache size.Finally,extensive simulation results are illustrated to show that the expression is a good approximation of the item sojourn time.展开更多
基金This work is supported by the National Natural Science Foundation of China under Grant No.61370091and No.61170200, Jiangsu Province Science and Technology Support Program （industry） Project under Grant No.BE2012179, Program Sponsored for Scientific Innovation Research of College Graduate in Jiangsu Province under Grant No. CXZZ12_0229.
文摘In this paper,we propose a novel spatial data index based on Hadoop:HQ-Tree.In HQ-Tree,we use PR QuadTrec to solve the problem of poor efficiency in parallel processing,which is caused by data insertion order and space overlapping.For the problem that HDFS cannot support random write,we propose an updating mechanism,called "Copy Write",to support the index update.Additionally,HQ-Tree employs a two-level index caching mechanism to reduce the cost of network transferring and I/O operations.Finally,we develop MapReduce-based algorithms,which are able to significantly enhance the efficiency of index creation and query.Experimental results demonstrate the effectiveness of our methods.
基金This work was supported by National Key Basic Research Program of China （973 Program） under Grant No. 2012CB315802 National Natural Science Foundation of China under Grant No. 61171102 and No. 61132001 Prospective Research on Future Networks of Jiangsu Future Networks Innovation institute under Grant No. BY2013095-4-01. Beijing Nova Program under Grant No.2008B50 and Beijing Higher Education Young Elite Teacher Project under Grant No.YETP0478.
文摘In-network caching and Interest packets aggregation are two important features of Content-Centric Networking(CCN).CCN routers can directly respond to the Interest request by Content Store(CS)and aggregate the same Interest packets by Pending Interest Table(PIT).In this way,most popular content requests will not reach the origin content server.Thus,content providers will be unaware of the actual usages of their contents in network.This new network paradigm presents content providers with unprecedented challenge.It will bring a great impact on existing mature business model of content providers,such as advertising revenue model based on hits amount.To leverage the advantages of CCN and the realistic business needs of content providers,we explore the hits-based content provisioning mechanism in CCN.The proposed approaches can avoid the unprecedented impact on content providers' existing business model and promote content providers to embrace the real deployment of CCN network.
基金Supported by the National Natural Science Foundation of China (No.60302004), the Australian Research Council (Grant LX0240468) and Natural Science Foun-dation of Hubei, China (No.2005ABA264).
文摘Due to large size and different popularity for different part of the video, most proxy caches for streaming medias cache only a part of the video. Thus, an accurate understanding on the internal popularity distribution of media objects in streaming applications is very important for the development of efficient cache mechanisms. This letter shows that the internal popularity of popular streaming media obeys a k-transformed Zipf-like distribution through analyzing two 6-month long traces recorded at different streaming video servers of an entertainment video-on-demand provider. This empirical model can be used to design an efficient cach- ing algorithm.
基金supported in part by the 973 Program under Grant No.2013CB329100in part by NSFC under Grant No.61422101,62171200,and 62132017+1 种基金in part by the Ph.D. Programs Foundation of MOE of China under Grant No.20130009110014in part by the Fundamental Research Funds for the Central Universities under Grant No.2016JBZ002
文摘Content-Centric Networking is a novel future network architecture that attracts increasing research interests in recent years. In-network caching has been regarded as a prominent feature of Content-Centric Networking since it is able to reduce the network traffic, alleviate the server bottleneck and decrease the user access latency. However, the CCN default caching scheme results in a high caching redundancy, causing an urgent need for an efficient caching scheme. To address this issue, we propose a novel implicit cooperative caching scheme to efficiently reduce the caching redundancy and improve the cache resources utilization. The simulation results show that our design achieves a higher hit ratio and a shorter cache hit distance in comparison with the other typical caching schemes.
文摘To reduce network redundancy,innetwork caching is considered in many future Internet architectures,such as Information Centric Networking.In in-network caching system,the item sojourn time of LRU(Least Recently Used) replacement policy is an important issue for two reasons:firstly,LRU is one of the most common used cache policy;secondly,item sojourn time is positively correlated to the hit probability,so this metric parameter could be useful to design the caching system.However,to the best of our knowledge,the sojourn time hasn't been studied theoretically so far.In this paper,we first model the LRU cache policy by Markov chain.Then an approximate closedform expression of the item expectation sojourn time is provided through the theory of stochastic service system,which is a function of the item request rates and cache size.Finally,extensive simulation results are illustrated to show that the expression is a good approximation of the item sojourn time.
