Content deleted Content added
No edit summary |
→History: minor fixes and improvements Tags: Mobile edit Mobile web edit Advanced mobile edit |
||
(41 intermediate revisions by 32 users not shown) | |||
Line 1:
{{one source |date=May 2024}}
The '''
Leslie Lamport, K. Mani Chandy: [https://research.microsoft.com/users/lamport/pubs/pubs.html#chandy ''Distributed Snapshots: Determining Global States of a Distributed System'']. In: ''ACM Transactions on Computer Systems 3''. Nr. 1, February 1985. ([http://lamport.azurewebsites.net/pubs/chandy.pdf PDF; 1 MB])</ref>
==History==
According to [http://research.microsoft.com/users/lamport/pubs/pubs.html#chandy Leslie Lamport's website],
==Definition==
The assumptions of the algorithm are as follows:
* There are no failures and all messages arrive intact and only once
* The communication channels are unidirectional and [[FIFO (computing and electronics)|FIFO]] ordered
* There is a communication path between any two processes in the system
* Any process may initiate the snapshot algorithm
Line 19:
Some of the assumptions of the algorithm can be facilitated using a more reliable communication protocol such as [[Internet protocol suite|TCP/IP]]. The algorithm can be adapted so that there could be multiple snapshots occurring simultaneously.
==
The
# The observer process (the process taking a snapshot):
## Saves its own local state
Line 27 ⟶ 28:
## Sends the observer process its own saved state
## Attaches the snapshot token to all subsequent messages (to help propagate the snapshot token)
#
From this, the observer builds up a complete snapshot: a saved state for each process and all messages “in the ether” are saved.
==References==
{{Reflist}}
{{DEFAULTSORT:Chandy-Lamport algorithm}}
[[Category:Distributed algorithms]]
|