1 / 22

LessLog A Logless File Replication Algorithm for Peer-to-Peer Distributed Systems

LessLog A Logless File Replication Algorithm for Peer-to-Peer Distributed Systems. Kuang-Li Huang Tai-Yi Huang Jerry C. Y. Chou Embedded Operating Systems (EOS) Lab http://eos.cs.nthu.edu.tw/ Department of Computer Science National Tsing Hua University, Taiwan. Outlines. Introduction

Download Presentation

LessLog A Logless File Replication Algorithm for Peer-to-Peer Distributed Systems

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. LessLogA Logless File Replication Algorithm for Peer-to-Peer Distributed Systems Kuang-Li Huang Tai-Yi Huang Jerry C. Y. Chou Embedded Operating Systems (EOS) Lab http://eos.cs.nthu.edu.tw/ Department of Computer Science National Tsing Hua University, Taiwan

  2. Outlines • Introduction • System model • Basic model • Advanced model • Fault-tolerant model • Experimental results • Conclusions and future work EOS Lab, National Tsing Hua University

  3. Related Work • Chord: A Scalable Peer-to-Peer Lookup Protocol for Internet Applications.IEEE/ACM Transactions on Networking, 2003 • Pastry: Scalable, Distributed Object Location and Routing for Large-Scale Peer-to-Peer Systems. IFIP/ACM International Conference on Distributed Systems Platforms (Middleware), 2001 • Tapestry: A Resilient Global-Scale Overlay for Service Deployment.IEEE Journal on Selected Areas in Communications, January 2004 • SCAN: A Dynamic, Scalable, and Efficient Content Distribution Network.International Conference on Pervasive Computing, 2002 EOS Lab, National Tsing Hua University

  4. LessLog Contributions • An efficient bitwise replication algorithm • Independent with underlying lookup protocols • Make no use of client-access logs • A complete set of file-access API • Fault-tolerant features • Self-organized for dynamic system change EOS Lab, National Tsing Hua University

  5. Basic System Model • There are totally Nlive nodes, N = 2m. • Each node is assigned a unique ID in [0, N-1] • Randomly assigned or any user-specified way • Such ID is called the PID of the node • Denoted by P(i) • A complete N-node binomial tree is built for each node • Totally N different physical trees EOS Lab, National Tsing Hua University

  6. 111 110 101 011 010 001 100 000 Virtual Tree • We use one unique virtual tree to construct each of N different physical trees VID EOS Lab, National Tsing Hua University

  7. complement Physical Tree for Node 3 3 3=011 key=100 111 110⊕100=010 2 1 7 101⊕100=001 110 101 011 0 6 010 001 100 4 000 5 EOS Lab, National Tsing Hua University

  8. 7 4 2 5 3 0 1 Physical Tree for Node 6 6 111 110 101 011 010 001 100 000 EOS Lab, National Tsing Hua University

  9. Properties of Virtual Tree • A node has i children nodes if its leftmost i bits in VID are all 1’s • The node of VID i has more or the same offspring nodes than the node of VID j if i > j • Given the PID of the root node in a lookup tree, we can do PID VID mapping for each node in the lookup tree EOS Lab, National Tsing Hua University

  10. 2 1 7 0 6 5 4 LessLog Tree Traversal 3 111 110 101 011 010 001 100 000 EOS Lab, National Tsing Hua University

  11. 2 1 7 0 6 5 4 connect toP(6) LessLog INSERT / GET API 3 111 110 101 011 010 001 100 000 askHashing(file)=3 EOS Lab, National Tsing Hua University

  12. LessLog REPLICATE API 3 111 110 101 011 010 001 100 000 Each replication reduces 50% load! EOS Lab, National Tsing Hua University

  13. Advanced System Model • There are totally N live nodes, N ≦ 2m. • Each node is assigned a unique ID in [0, 2m-1] • Randomly assigned or any user-specified way • Denoted by P(i) • Live nodes and dead nodes • A complete 2m-node binomial tree is built for each node • Totally 2m different physical trees EOS Lab, National Tsing Hua University

  14. AdvancedLessLog INSERT / GET API The 1st live node with largest VID 3 111 2 1 7 110 101 011 0 6 5 010 001 100 4 000 askHashing(file)=3 EOS Lab, National Tsing Hua University

  15. 7 0 6 5 4 Overloaded Node with Largest VID 3 111 2 1 110 101 011 100 010 001 100 011 010 000 Sorted by VID 001 EOS Lab, National Tsing Hua University

  16. 6 7 111 111 7 4 6 2 5 3 110 101 011 110 101 011 5 3 2 0 1 4 001 100 010 001 100 010 0 100 000 Self-organized LessLog • Joining (Leaving) a node • Check whether it is the largest VID • Ex.P(6) join, P(1), P(4) are dead nodes 1 EOS Lab, National Tsing Hua University

  17. 1 1 0 1 0 0 1 0 Fault-Tolerant LessLog • Files can always be accessed • LessLog stores each file in 2bnodes,b≦m • 2b identical subtrees 3 111 2 1 7 110 101 011 0 6 5 100 010 001 4 000 EOS Lab, National Tsing Hua University

  18. 22 identical subtrees 3 1111 1 7 11 2 1101 1011 0111 1110 15 0 6 10 5 9 0101 1001 0011 1100 1010 0110 14 4 8 13 0100 1000 0010 0001 12 0000 EOS Lab, National Tsing Hua University

  19. Experimental Results • Compared with log-based and random algorithms at an evenly-distributed model replicas incoming requests/1000 EOS Lab, National Tsing Hua University

  20. Experimental Results • Compared with log-based and random algorithms at an locality model replicas incoming requests/1000 EOS Lab, National Tsing Hua University

  21. Conclusions and Future Work • An efficient bitwise replication algorithm • Built on top of any existing lookup protocol • Make no use of client-access logs • Significant performance improvement • Fault-tolerant & self-organized futures • Concurrent joins / leaves / failures • Under implementation at PlanetNet systems • Real performance numbers EOS Lab, National Tsing Hua University

  22. Thank You EOS Lab, National Tsing Hua University

More Related