/[gzz]/gzz/Documentation/misc/hemppah-progradu/masterthesis.tex
ViewVC logotype

Diff of /gzz/Documentation/misc/hemppah-progradu/masterthesis.tex

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.93 by hemppah, Thu Feb 27 13:52:24 2003 UTC revision 1.94 by hemppah, Thu Feb 27 14:09:19 2003 UTC
# Line 386  value of $n$ varies among approaches. Ag Line 386  value of $n$ varies among approaches. Ag
386  to implement identifier space.  to implement identifier space.
387    
388  To store data into tightly structured overlay, each application-specific  To store data into tightly structured overlay, each application-specific
389  key is \emph{mapped} by the overlay to a existing peer in the overlay. Each  key is \emph{mapped} by the overlay to a existing peer in the overlay. Thus, tightly
390  peer in the structured overlay maintains a \emph{routing table}, which consists  structured overlay assigns a subset of all possible keys to every participating peer.
391  of identifiers and IP addresses of other peers in the overlay. These are peer's  Furtermore, each peer in the structured overlay maintains a \emph{routing table},
392  neighbors in the overlay network. Figure \ref{fig:structured_hashing}  which consists of identifiers and IP addresses of other peers in the overlay.
393    These are peer's neighbors in the overlay network. Figure \ref{fig:structured_hashing}
394  illustrates the process of data to key mapping in tightly strucuted overlays.  illustrates the process of data to key mapping in tightly strucuted overlays.
395    
396  \begin{figure}  \begin{figure}
# Line 419  support for a distance function. Finally Line 420  support for a distance function. Finally
420  must be constructed and maintained adaptively.  must be constructed and maintained adaptively.
421    
422  Currently, all proposed tightly structured overlays provide at least  Currently, all proposed tightly structured overlays provide at least
423  poly-logaritmical data lookup operations. However, there are some key  poly--logaritmical data lookup operations. However, there are some key
424  differences in routing algoritms. For example, Chord, Skip graphs and  differences in routing algoritms. For example, Chord, Skip graphs and
425  Skipnet maintain a local data structure which resembles skip lists \cite{78977}.  Skipnet maintain a local data structure which resembles skip lists \cite{78977}.
426  In figure \ref{fig:structured_query}, we present overview of Chord's lookup process.  In figure \ref{fig:structured_query}, we present overview of Chord's lookup process.
427  On the left side of Chord's lookup process, we show the same data lookup process  On the left side of Chord's lookup process, we show the same data lookup process
428  as binary-tree abstraction.  We can notice, that in each step, the distance between  as binary-tree abstraction.  We can notice, that in each step, the distance between
429  the query originator and the target in both methods is halved. Thus, the  the query originator and the target in both methods is halved. Thus, the
430  locarithmic efficiency. Kademlia, Pastry and Tapestry uses balanced tree-like  locarithmic efficiency.
431    
432    Kademlia, Pastry and Tapestry uses balanced tree-like
433  data structures. Figure \ref{fig:kademlia_lookup} shows the process of Kademlia  data structures. Figure \ref{fig:kademlia_lookup} shows the process of Kademlia
434  data lookup. Viceroy maintains a butterfly data structure, which requires  data lookup. Viceroy maintains a butterfly data structure, which requires
435  only constant number of neighbor peers while providing $O(\log{n})$ data lookup  only constant number of neighbor peers while providing $O(\log{n})$ data lookup
436  efficiency. Koorde, recent modification of Chord, uses de Bruijn graphs to maintain  efficiency. Koorde, recent modification of Chord, uses de Bruijn graphs to maintain
437  local routing tables. Koorde requires each peer to have only about two links to other  local routing tables. Koorde requires each peer to have only about two links to other
438  peers to to provide $O(\log{n})$ performance.  peers to to provide $O(\log{n})$ performance. Peernet
439    
440  \begin{figure}  \begin{figure}
441  \centering  \centering
# Line 450  peers to to provide $O(\log{n})$ perform Line 453  peers to to provide $O(\log{n})$ perform
453  \end{figure}  \end{figure}
454    
455    
456    Chord example ?
457    
458    abstraction: DHT, DOLR, multicast/anycast
459    
460    
461  -service is data block, node/peer is a physical computer  -service is data block, node/peer is a physical computer
462  -*servers* self-organize towards a lookup network  -*servers* self-organize towards a lookup network
# Line 532  where $difference(p,p_neighbor)= close$, Line 539  where $difference(p,p_neighbor)= close$,
539    
540  \subsection{Protocols}  \subsection{Protocols}
541    
 Measures:  
542    
 degree:  
   
 hop count:  
   
 fault-tolerance  
   
 maintenance overhead  
   
 load balance  
543    
544    
545    
# Line 620  load balance Line 617  load balance
617    
618  \section{Summary}  \section{Summary}
619    
620    Measures:
621    
622    degree:
623    
624    hop count:
625    
626    fault-tolerance
627    
628    maintenance overhead
629    
630    load balance
631    
632    
633  \subsection{Differences}  \subsection{Differences}
634    

Legend:
Removed from v.1.93  
changed lines
  Added in v.1.94

savannah-hackers-public@gnu.org
ViewVC Help
Powered by ViewVC 1.1.26