Presentation is loading. Please wait.

Presentation is loading. Please wait.

Scalable Name Lookup in NDN Using Effective Name Component Encoding Yi Wang, Keqiang He, Huichen Dai, Wei Meng, Junchen Jiang, Bin Liu, Yan Chen.

Similar presentations


Presentation on theme: "Scalable Name Lookup in NDN Using Effective Name Component Encoding Yi Wang, Keqiang He, Huichen Dai, Wei Meng, Junchen Jiang, Bin Liu, Yan Chen."— Presentation transcript:

1 Scalable Name Lookup in NDN Using Effective Name Component Encoding Yi Wang, Keqiang He, Huichen Dai, Wei Meng, Junchen Jiang, Bin Liu, Yan Chen

2 Parallel Name Lookup for NDN Outline 1. Named Data Networking (NDN) Introduction 2. Name Lookup in NDN 3. Name Component Encoding (NCE) 4. Analysis 5. Experimental Results 6. Conclusion

3 Parallel Name Lookup for NDN NDN Introduction Named Data Networking (NDN) Named Data Networking is proposed recently as the clean-slate network architecture for future Internet, which no longer concentrates on where the information is located, but what the information (content) is needed. NDN uses names to identify every piece of contents instead of IP addresses for hardware devices attached to IP network.

4 Parallel Name Lookup for NDN NDN Introduction Naming in NDN An NDN name is hierarchically structured and composed of explicitly delimited components Interest and Data Packets in NDN /com/google/maps comgooglemaps

5 Parallel Name Lookup for NDN NDN Introduction Packet Forwarding Process Interest Packet Data Packet Client Content Provider Dst Src IP Packet Content Store FIB

6 Parallel Name Lookup for NDN NDN Introduction Packet Forwarding Process

7 Parallel Name Lookup for NDN Outline 1. Named Data Networking (NDN) Introduction 2. Name Lookup in NDN 3. Name Component Encoding (NCE) 4. Analysis 5. Experimental Results 6. Conclusion

8 Parallel Name Lookup for NDN Name Lookup in NDN The challenges of name lookup as below: Variable length name: unlimited components number and unfixed components length Longest name prefix matching: aggregate prefixes to reduce the total number of prefixes in FIB Interest Packet and Data Packet has different lookup processes The large-scale name prefix set Frequently update

9 Parallel Name Lookup for NDN Name Lookup in NDN Name lookup at component granularity / com / yahoo / news / com / yahoo / music / new / com / google / news / com / google / cn / com / sina / news / cn / com / sina / mail / cn / yahoo / news 1 1 2 2 3 3 4 4 5 5 6 6 7 7 com cn 8 8 9 9 A A B B C C yahoo google yahoo com sina news music news D D E E F F new news mail level-1level-5level-2level-4level-3 Name Prefix Trie (NPT )

10 Parallel Name Lookup for NDN Outline 1. Named Data Networking (NDN) Introduction 2. Name Lookup in NDN 3. Name Component Encoding (NCE) 4. Analysis 5. Experimental Results 6. Conclusion

11 Parallel Name Lookup for NDN NCE Algorithm 1 1 2 2 9 9 3 3 7 7 A A D D com cn 4 4 5 5 8 8 B B E E yahoo google baidu google maps map news maps 6 6 uk level-1level-5level-2level-4level-3 C C sina NamePointer /com/yahoo… /com/yahoo/news… /com/yahoo/maps/uk… /com/google… /com/google/maps… /cn/google/maps… /cn/sina… /cn/baidu… /cn/baidu/map…

12 NameCodingPorts /com/yahoo/1/11 /com/yahoo/news/1/1/11 /com/yahoo/maps/uk/1/1/3/12 /com/google/1/42 /com/google/maps/1/4/31, 2 /cn/google/maps/2/4/33 /cn/sina/2/32, 3 /cn/baidu/2/14 /cn/baidu/map/2/1/14 1 1 2 2 3 3 4 4 5 5 6 6 8 8 com,1 cn,2 9 9 A A B B C C D D yahoo,1 google, 4 baidu,1 google,4 maps,3 map,1 news,1 maps,3 E E uk,1 level-1level-5level-2level-4level-3 7 7 sina,3 00014001800100040007800500094004000A000B000D000E000F0010 Base: (hex) Transition_1: 2122131300110000 0231910411720145693 21411 045813 313413 0876012 Transition_2: Transition_4: # of Transitions Ports List Pointer Transition 1 2 3 4 5 6 7

13 Parallel Name Lookup for NDN NCE Algorithm 1 1 2 2 3 3 5 5 c 4 4 o n m 1c2on1m00 020350412 13689 Base: Transition: 1 2 3 4 5 6 CodeStates List 12.. 29.. Character Trie for Components: com cn

14 Parallel Name Lookup for NDN Outline 1. Named Data Networking (NDN) Introduction 2. Name Lookup in NDN 3. Name Component Encoding (NCE) 4. Analysis 5. Experimental Results 6. Conclusion

15 Parallel Name Lookup for NDN Analysis Memory Character Trie: α=8, β=9 Name Component Trie: α=9, β=5

16 Parallel Name Lookup for NDN Analysis In summary, compared with NCT, NCE utilizes the following three parts to reduce storage overhead. NCE uses State Transition Arrays to construct the NCT, and the memory cost can be reduced at least save 17.64%; Code Allocation Mechanism reduces the number of components by merging the Original Collision Set at the same level; NCE stores the transitions in different sizes of Transition Arrays. Compared with the method that uses Transition only, it can reduce the memory overhead further.

17 Parallel Name Lookup for NDN Analysis In NCE, the longest name prefix matching contains two Steps: So, a name lookup has: 2) looks up codes in ENPT-STA: 1) finds the components corresponding codes in CCT-STA: If there are P parallel code lookup modules, the complexity can be reduced to:

18 Parallel Name Lookup for NDN Analysis Compared with character trie, NCE can gains:

19 Parallel Name Lookup for NDN Outline 1. Named Data Networking (NDN) Introduction 2. Name Lookup in NDN 3. Name Component Encoding (NCE) 4. Analysis 5. Experimental Results 6. Conclusion

20 Parallel Name Lookup for NDN Experimental Results Number of Domains with different components number:

21 Parallel Name Lookup for NDN Experimental Results Comparison of memory usage:

22 Parallel Name Lookup for NDN Experimental Results The number of different components and codes, and the compression ratio of Code Allocate Mechanism on DMOZ dataset:

23 Parallel Name Lookup for NDN Experimental Results Number of Entries for Transition1, Transition2 and Transition4 on DMOZ dataset:

24 Parallel Name Lookup for NDN Experimental Results The Memory Cost of NCE and NCT on DMOZ dataset:

25 Parallel Name Lookup for NDN Experimental Results Comparison of NCT and NCEs processing performance:

26 Parallel Name Lookup for NDN Experimental Results NCEs Average Lookup Time (When the Number of Parallel CCT lookup modules is 3):

27 Parallel Name Lookup for NDN Experimental Results The relationship between NCEs average lookuptime and the number of parallel CCT lookup modules

28 Parallel Name Lookup for NDN Experimental Results The relationship between NCEs speedup and the number of parallel CCT lookup modules

29 Parallel Name Lookup for NDN Experimental Results The relationship between NCEs packet delay and the number of parallel CCT lookup modules

30 Parallel Name Lookup for NDN Outline 1. Named Data Networking (NDN) Introduction 2. Name Lookup in NDN 3. Name Component Encoding (NCE) 4. Analysis 5. Experimental Results 6. Conclusion

31 Parallel Name Lookup for NDN Conclusion Proposed an effective Name Components Encoding approach: Code Allocation Mechanism State Transition Array Both theoretical analysis and experiments on real domain sets demonstrate that NCE could effectively reduce the memory cost while guaranteeing high-speed of longest name prefix lookup.

32 Parallel Name Lookup for NDN Thank you! Q & A


Download ppt "Scalable Name Lookup in NDN Using Effective Name Component Encoding Yi Wang, Keqiang He, Huichen Dai, Wei Meng, Junchen Jiang, Bin Liu, Yan Chen."

Similar presentations


Ads by Google