<address id="xrvvb"><listing id="xrvvb"></listing></address>

          <form id="xrvvb"></form>

          <address id="xrvvb"></address>

          華中科技大學學報(自然科學版) 2020, Vol. 48 Issue (9): 50-56 DOI10.13245/j.hust.200909

          欄目:計算機與控制工程
          基于雙向跳表的命名數據網內容存儲池
          劉開華 , 周美麗 , 李 卓
          天津大學微電子學院,河北 天津 300072
          摘要 針對命名數據網中內容存儲池須要快速檢索差異化名稱數據、高效存儲數據信息和有效支持緩存替換策略的設計需求,提出了一種基于雙向跳表的命名數據網內容存儲池的存儲結構學習樹-雙向跳表(LT-DSL).該存儲結構由部署于片上的學習樹和部署于片下的雙向跳表(DSL)兩種數據結構組成.LT-DSL通過重新設計索引數據結構來學習索引內容在存儲器中的分布情況,實現了數據的均勻映射,提升了存儲效率,降低了存儲開銷;同時,通過采用全新的名稱數據檢索算法來快速檢索名稱數據,LT-DSL有效支持了緩存替換策略,加速了數據處理過程.實驗結果表明:相比其他基于跳表的內容存儲池方案,LT-DSL在有效支持緩存替換策略和滿足網絡丟包率低于1%的前提下,進一步提升了數據檢索速度,降低了整體存儲開銷.
          關鍵詞 命名數據網 ;轉發平面 ;內容存儲池 ;神經網絡 ;跳表 ;名稱數據檢索
          Content store via double skip list for named data networking
          LIU Kaihua , ZHOU Meili , LI Zhuo
          School of Microelectronics,Tianjin University,Tianjin 300072,China
          Abstract According to the requirements which include performing fast name lookup,providing high capacity,and supporting cache replacements are required in content store for improving the overall forwarding performance of named data networking (NDN),an enhanced content store named learning tree-double skip list (LT-DSL) was proposed based on double skip list (DSL).LT-DSL consists of two parts:on-chip-based learning tree and off-chip-based DSL.LT-DSL can learn the distribution characteristic of data indexed in static memory by redesigning the index,which achieves more uniform mapping,improves the memory utilization,and reduces memory consumption.Besides,LT-DSL can realize the fast name lookup and support cache replacements by utilizing the new name lookup algorithms.The experiment results demonstrate that under the premise of supporting the cache replacements,LT-DSL can promote the lookup speed,and further minimize the memory consumption with the probability of false positive under 1%.
          Keywords named data networking ; forwarding plane ; content store ; neural network ; skip list ; name lookup
          基金資助國家自然科學基金資助項目(61602346)

          中圖分類號TP393.0
          文獻標志碼A
          文章編號1671-4512(2020)09-0050-07
          參考文獻
          [1] ZHANG L,ESTRIN D,BURKE J,et al.Named data networking (NDN) project NDN-0001[EB/OL].Palo Alto:Xerox Palo Alto Research Center.[2019-01-20].http://named-data.net/.
          [2] 宋其翰,吳軍,馬嚴,等.Path-LCE:基于LCE和路徑協同的NDN緩存策略[J].華中科技大學學報(自然科學版),2016,44(S1):62-66.
          [3] LI Z,XU Y,ZHANG B,et al.Packet forwarding in named data networking requirements and survey of solutions[J].IEEE Communications Surveys & Tutorials,2018,21(2):1950-1987.
          [4] AFANASYEV A,SHI J,ZHANG B,et al.Nfd developer's guide NDN-0021[EB/OL].Palo Alto:Xerox Palo Alto Research Center.[2019-10-20].http://named-data. net/.
          [5] YUAN H.Data structures and algorithms for scalable NDN forwarding[D].Washington:Engineering & Applied Science,Washington University in St.Louis,2015.
          [6] THOMAS Y,XYLOMENOS G,TSILOPOULOS C,et al.Object-oriented packet caching for ICN[C]// Proc of 2nd ACM Conference on Information-Centric Network- ing.New York:ACM,2015:89-98.
          [7] ZHANG R,LIU J,HUANG T,et al.Adaptive compression trie based bloom filter:request filter for NDN content store[J].IEEE Access,2017,5:23647-23656.
          [8] HOU R,ZHANG L,WU T,et al.Bloom-filter-based request node collaboration caching for named data networking[J].Cluster Computing,2018,2018(4):1-12.
          [9] YUAN H,SONG T,CROWLEY P.Scalable NDN forwarding:concepts,issues and principles[C]// Proc of 21st International Conference on Computer Communi- cations and Networks.Piscataway:IEEE,2012:1-9.
          [10] PAN T,HUANG T,LIU J,et al.Fast content store lookup using locality-aware skip list in content-centric networks [C]// Proc of Computer Communications Workshops (INFOCOM WKSHPS).Piscataway:IEEE,2016:187-192.
          [11] 潘恬,黃韜,張雪貝.基于局部性原理跳表的內容路由器緩存快速檢索機制[J].計算機學報,2018,41(9):2029-2043.
          [12] YAN L,LI Z,LIU K.Learning tree:neural network-based index for NDN forwarding plane[C]// Proc of ACM SIGCOMM 2019 Conference Posters and Demos.New York:ACM,2019:63-65.
          [13] SHI J,LIANG T,WU H,et al.Ndn-NIC:name-based filtering on network interface card[C]// Proc of 3rd ACM Conference on Information-Centric Networking.New York:ACM,2016:40-49.
          [14] FAYAZBAKHSH S,LIN Y,TOOTOONCHIAN A,et al.Less pain,most of the gain:incrementally deployable ICN[C]// Proc of ACM SIGCOMM Computer Communi- cation Review.New York:ACM,2013:147-158.
          [15] RUMELHART D E,HINTON G E,WILLIAMS R J.Learning representations by back-propagating errors [J].Nature,1986,323(6088):533-536.
          [16] PUGH W.Skip lists:a probabilistic alternative to balanced trees[C]// Proc of Workshop on Algorithms & Data Structures.New York:ACM,1990:668-676.
          [17] KIRSCH A,MITZENMACHER M,VARGHESE G.Hash-based techniques for high-speed packet processing[C]// Proc of Algorithms for Next Generation Networks.Ber-lin:Springer-Verlag,2010:181-218.
          [18] LI Z,XU Y,LIU K,et al.5G with B-MaFIB based named data networking[J].IEEE Access,2018,6(1):30501-30507.
          [19] PERINO D,VARVELLO M.A reality check for content centric networking[C]// Proc of ACM SIGCOMM Work- Shop on Information-Centric Networking.New York:ACM,2011:44-49.
          [20] SONG T,YUAN H,CROWLEY P,et al.Scalable name-based packet forwarding:from millions to billions[C]// Proc of 2nd International Conference on Information-Centric Networking.New York:ACM,2015:19-28.
          文獻來源
          劉開華, 周美麗, 李 卓. 基于雙向跳表的命名數據網內容存儲池[J]. 華中科技大學學報(自然科學版), 2020, 48(9): 50-56
          DOI:10.13245/j.hust.200909
          澳门盘口