Abstract:
We present the design, implementation, and evaluation of a fully distributed directory service for Farsite, a logically centralized file system that is physically implemented on a loosely coupled network of desktop computers. Prior to this work, the Farsite system included distributed mechanisms for file content but centralized mechanisms for file metadata. Our distributed directory service introduces tree-structured file identifiers that support dynamically partitioning metadata at arbitrary granularity, recursive path leases for scalably maintaining name-space consistency, and a protocol for consistently performing operations on files managed by separate machines. It also mitigates metadata hotspots via file-field leases and the new mechanism of disjunctive leases. We experimentally show that Farsite can dynamically partition file-system metadata while maintaining full file-system semantics.
Link to the full paper:
http://www.cse.buffalo.edu/faculty/tkosar/cse710_spring13/papers/farsite.pdf
http://www.cse.buffalo.edu/faculty/tkosar/cse710_spring13/papers/farsite.pdf
If the server delegates its metadata to another machine in the system, how could client still work properly? Will the server send some redirection message to the client or make other changes?
ReplyDeleteHere file identifiers have a tree structure, which supports arbitrarily fine-grained partitioning and name space is always kept consistent(recursive path leases). Each server manages identifiers beginning with a specified prefix, except for those it has explicitly delegated away to other servers. At any moment, some portion of each file identifier determines which server manages the file. Hence there is no need of redirection message.
DeleteThe Byzantine Fault Tolerant Protocol is a disadvantage on performance. How does this affect the scalability of the system?
ReplyDeleteFarsite’s metadata service was at first centralized on a single BFT group of machines, which could not scale to the normal file-system metadata loads of the ~10^5 desktop computers as it had throughput of a single machine, but with automated load balancing and distributed metadata, there is no performace problem in scaling.
Delete