1 / 12

Lock Services in Distributed File Systems

Lock Services in Distributed File Systems. Shaan Mahbubani Anshuman Gupta Ravi Vijay Anup Tapadia. UCSD CSE 221 Operating Systems - Winter 07. Distributed File System Lock Service . Synchronization Consistency Access control Shared access. Farsite - Overview. Goals:

benjamin
Download Presentation

Lock Services in Distributed File Systems

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Lock Services in Distributed File Systems Shaan Mahbubani Anshuman Gupta Ravi Vijay Anup Tapadia UCSD CSE 221 Operating Systems - Winter 07

  2. Distributed File System Lock Service • Synchronization • Consistency • Access control • Shared access

  3. Farsite - Overview Goals: • Meta-data partitioning • Namespace consistency • Scalable Design: • Recursive path leases • File-field leases • Disjunctive leases

  4. Chubby - Overview Goals: • Coarse-grained synchronization • Primary election Design: • High capacity • Server, client library • Sessions, events, client caching

  5. Analysis - Farsite • Fully functional rename • Identifier based partitioning • Fine grained synchronization: leases • Logical atomicity • Optimizations: • Recursive leases • Lazy updates • No fault recovery

  6. Analysis - Chubby • Low volume storage • Flat access control lists • Lock service in lieu of consensus service • File handles and advisory locks • Event notification service • KeepAlive piggybacking • Client side caching

  7. Comparison

  8. More Comparison

  9. Even More Comparison

  10. Questions • Farsite • Scalability • Lease overhead penalty • Network robustness • Effect of malicious machines • Time to recovery • Chubby • API problems • Throughput bottlenecks due to keepalives • System capacity

  11. Conclusion • Farsite locking • Distributed directory service avoids hotspots • Fine grained locking avoids false sharing • No fault tolerance • Chubby locking • Lock service is better than consensus service • Coarse grain locking more useful • More reliable across failovers

  12. Thank You ! Questions ?

More Related