59 |
-Employs randomness to achieve both load distribution and routing |
-Employs randomness to achieve both load distribution and routing |
60 |
locality |
locality |
61 |
|
|
62 |
|
2.4. Pastry |
63 |
|
|
64 |
|
-Project of Microsoft reasearch ("interesting") |
65 |
|
-Peer-to-Peer anonymous storage |
66 |
|
-Location protocol sharing many similarities with Tapestry |
67 |
|
-Objects are replicated without any permission of the owner |
68 |
|
|
69 |
|
|
70 |
|
|
71 |
2.3. Gnutella |
2.3. Gnutella |
72 |
|
|
124 |
-Search mechanisms are unscalable: a great amount of loads are generated |
-Search mechanisms are unscalable: a great amount of loads are generated |
125 |
on the network during search requests |
on the network during search requests |
126 |
|
|
127 |
|
Miscellaneous notes: |
128 |
|
|
129 |
|
-Object's popularity: studies have shown that Napster, Gnutella and Web |
130 |
|
queries follow Zipf-like distributions (1/2, 1/3, 1/4 etc.) [3, 4] |
131 |
|
|
132 |
|
|
133 |
Bibliography: |
Bibliography: |
134 |
|
|
137 |
|
|
138 |
[2] Lv, Qin, Cao Pei, Cohen Edith, Li Kai, Shenker Scott, Search and |
[2] Lv, Qin, Cao Pei, Cohen Edith, Li Kai, Shenker Scott, Search and |
139 |
Replication in Unstructured Peer-to-Peer Networks |
Replication in Unstructured Peer-to-Peer Networks |
140 |
|
|
141 |
|
[3] Aiello W., A Random graph model for massive graphs |
142 |
|
|
143 |
|
[4] Sripanidkulchai K., The popularity of gnutella queries and its |
144 |
|
implications on scalability |