/[storm]/storm/doc/pegboard/attacking_gisp--hemppah/peg.rst
ViewVC logotype

Diff of /storm/doc/pegboard/attacking_gisp--hemppah/peg.rst

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

revision 1.33 by hemppah, Fri Jun 13 12:42:12 2003 UTC revision 1.34 by hemppah, Fri Jun 13 13:12:32 2003 UTC
# Line 237  Hypothesis: Line 237  Hypothesis:
237    to maintain hostile peers in their routing tables while still hostile    to maintain hostile peers in their routing tables while still hostile
238    peers are able to cause problems to the network's operation. "SEEN", "RESULT"    peers are able to cause problems to the network's operation. "SEEN", "RESULT"
239    and "QUERY" messages are part of GISP-3.4 specification. Also,    and "QUERY" messages are part of GISP-3.4 specification. Also,
240    a hostile peer is able to use this "dumb" behaviour with other GISP-3.4    a hostile peer is able to use this "dumb" behaviour with the other messages
241    messages.    of GISP-3.4 specification.
242          
243  - In the GISP overlay, the average routing hop length is proportional to  - In the GISP overlay, the average routing hop length is proportional to
244    O(log n).    O(log n).
245        
# Line 248  Hypothesis: Line 248  Hypothesis:
248    probability of routing successfully "QUERY" message between two regular    probability of routing successfully "QUERY" message between two regular
249    peers is (1-f)^h-1, where h is the routing hops used in the overlay.    peers is (1-f)^h-1, where h is the routing hops used in the overlay.
250    
   
251        
252        
253  - - -  - - -
254    
255  - It is expected that (popular) data items can be found with fewer hops in  - It is expected that (popular) data items can be found with fewer hops in
256    *some cases* in GISP network w.r.t Chord network, since GISP extends Chord's    *some cases* in GISP network w.r.t Chord network, since GISP extends Chord's
257    routing table to have more information as a cache    routing table to have more information as a cache
258        
259  .. more specific, what effects cache has, formula ?  .. more specific, what effects cache has, distributions, formula ?
   
     
 - GISP's lookup failure rate increases linearly with the fraction of "dumb"  
   peers. More specifically, fraction f of lookups fail after a randomly  
   selected fraction of f peers become "dumb" in a static network  
     
 .. make "dumb"  
260        
261  - It is expected that fraction f of all network traffic k is "lost" in  - It is expected that fraction f of all network traffic k is "lost" in
262    a static GISP network and therefore may have negative influence to network's    a static GISP network and therefore may have negative influence to network's
# Line 276  Hypothesis: Line 269  Hypothesis:
269  If fraction f of the peers are dumb in a static network with 10 - 10000  If fraction f of the peers are dumb in a static network with 10 - 10000
270    peer and fraction f of the lookups, then fraction of the lookups will    peer and fraction f of the lookups, then fraction of the lookups will
271    fail at all network sizes    fail at all network sizes
272      
273    - GISP's lookup failure rate increases linearly in a static network with the
274      fraction of "dumb" peers when a "dumb" peer do not provide a "RESULT"
275      message for a "QUERY" message. More specifically, when randomly selected
276      fraction of f all peers become "dumb", same fraction f of all lookups will
277      fail.  
278    
279        
280  .. No, you should not say "expect similar" - it's far too vague.  .. No, you should not say "expect similar" - it's far too vague.

Legend:
Removed from v.1.33  
changed lines
  Added in v.1.34

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