[Gzz-commits] gzz/Documentation/misc/hemppah-progradu existin...

[Hermanni Hyytiälä]

CVSROOT:        /cvsroot/gzz
Module name:    gzz
Changes by:     Hermanni Hyytiälä <address@hidden>      02/11/11 08:19:26

Modified files:
        Documentation/misc/hemppah-progradu: existing_systems_overview 

Log message:
        Overview of Coral p2p system (very interesting !)

CVSWeb URLs:
http://savannah.gnu.org/cgi-bin/viewcvs/gzz/gzz/Documentation/misc/hemppah-progradu/existing_systems_overview.diff?tr1=1.6&tr2=1.7&r1=text&r2=text

Patches:
Index: gzz/Documentation/misc/hemppah-progradu/existing_systems_overview
diff -u gzz/Documentation/misc/hemppah-progradu/existing_systems_overview:1.6 
gzz/Documentation/misc/hemppah-progradu/existing_systems_overview:1.7
--- gzz/Documentation/misc/hemppah-progradu/existing_systems_overview:1.6       
Thu Nov  7 03:32:03 2002
+++ gzz/Documentation/misc/hemppah-progradu/existing_systems_overview   Mon Nov 
11 08:19:26 2002
@@ -136,6 +136,20 @@
 lookup substantially. The above enhancment cannot be implemented in a natural 
way in any of the well-known DHTs 
 (distributed hash tables) such as Chord, Pastry, Tapestry, etc. since the 
design of their routing tables allows 
 for no such flexibility, and hence they would require additional data 
structures to accommodate temporary contacts.
+
+2.6 Coral
+Coral is a peer-to-peer content distribution system that creates 
self-organizing clusters of nodes that fetch 
+information from each other to avoid communicating with more distant or 
heavily-loaded servers. Coral is 
+designed to index local data, such as the files in a node's web cache, to 
ensure that content is replicated 
+exactly in proportion to its use. Coral uses two novel mechanisms to achieve 
scalability and high performance. 
+First, a new abstraction called a distributed sloppy hash table (DSHT) lets 
nodes locate nearby copies of a 
+file without causing hot spots, regardless of its popularity. Second, based on 
the DSHT interface, a 
+decentralized clustering algorithm allows nodes to can find each other and 
form clusters of varying network 
+diameters.
+
+Efficient peer-to-peer lookup based on distributed tries.
+
+hemppah: This is very interesting approach (tuomasl et all, comments ?)