by J. Kistler et al., SOSP 1991
Abstract:
Disconnected operation is a mode of operation that enables a client to continue accessing critical data during temporary failures of a shared data repository. An important, though not exclusive, application of disconnected operation is in supporting portable computers. In this paper, we show that disconnected operation is feasible, efficient and usable by describing its design and imple- mentation in the Coda File System. The central idea behind our work is that caching of data, now widely used for performance, can also be exploited to improve availability.
Abstract:
Disconnected operation is a mode of operation that enables a client to continue accessing critical data during temporary failures of a shared data repository. An important, though not exclusive, application of disconnected operation is in supporting portable computers. In this paper, we show that disconnected operation is feasible, efficient and usable by describing its design and imple- mentation in the Coda File System. The central idea behind our work is that caching of data, now widely used for performance, can also be exploited to improve availability.
Link to the full paper:
http://www.cse.buffalo.edu/faculty/tkosar/cse710_spring13/papers/coda.pdf
Idea represented in the paper is that user can disconnect and reconnect without having to deal with issues, but if there are changes concurrently, can they be overcome or do they require manual intervention?
ReplyDeleteWriter says they made venus a user level process, but I am not clear about the support for their decision. They said complexity of venus - Is it a design complexity, or functional complexity.
ReplyDeleteCan you elaborate on the three potential strategies Venus uses to handle the resource exhaustion issue during emulation?
ReplyDeleteDoes coda have any load balancing mechanism among the replica servers? If yes, what is the strategy used?
ReplyDeleteHow does CODA implement secure communication between the servers and the clients? Are there any security mechanisms present in the client system such as authentication to prevent unauthoried access to cached files?
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteIt is said that the modifications in Coda are propagated to all the AVSG sites and eventually to the missing VSG sites. The conflict resolution discussed in section 4.5.2 says that the conflict is resolved at phase 2 of replay. However, only the AVSG sites are discussed about in the replay algorithm. What happens to the missing VSG sites? Are conflicts in different network partitions addressed here?
ReplyDeleteAs a future scope, it is proposed that CODA can be extended to wireless technologies for masking weak signal/connectivity. With the current trend of wide usage of wireless technology, how does latest version of CODA support this?
ReplyDeleteCan CODA be practically deployed over WANs? An example could be on servers for CDNs or mirroring servers where the servers also act like CODA clients to a centrally located repository and serve all the FTP/HTTP requests through the locally held cache..
ReplyDelete