NFSv4 Replication for Grid Computing
Download as PPT, PDF1 like1,082 views
The document discusses the challenges and solutions in implementing consistent replication for grid computing, focusing on ensuring efficient and reliable access to distributed data. It introduces a replication control protocol based on NFSv4 for managing file replication, emphasizing performance and fine-grained control. The conclusions highlight that implementing consistent mutable replication in large-scale systems is both feasible and enhances overall system utilization.
NFSv4 Replication for Grid Computing
- 1. NFSv4 Replication for Grid Computing Peter Honeyman Center for Information Technology Integration University of Michigan, Ann Arbor
- 2. Acknowledgements Joint work with Jiaying Zhang UM CSE doctoral candidate Defending later this month Partially supported by NSF/NMI GridNFS DOE/SciDAC Petascale Data Storage Institute Network Appliance, Inc. IBM ARC
- 3. Outline Background Consistent replication Fine-grained replication control Hierarchical replication control Evaluation Durability revisited NEW! Conclusion SKIP SKIP SKIP SKIP SKIP SKIP SKIP SKIP
- 4. Grid computing Emerging global scientific collaborations require access to widely distributed data that is reliable, efficient, and convenient SKIP SKIP SKIP Grid Computing
- 5. GridFTP Advantages Automatic negotiation of TCP options Parallel data transfer Integrated Grid security Easy to install and support across a broad range of platforms Drawbacks Data sharing requires manual synchronization SKIP SKIP SKIP
- 6. NFSv4 Advantages Traditional, well-understood file system semantics Supports multiple security mechanisms Close-to-open consistency Reader is is guaranteed to see data written by the last writer to close the file Drawbacks Wide-area performance SKIP SKIP SKIP
- 7. NFSv4.r Research prototype developed at CITI Replicated file system build on NFSv4 Server-to-server replication control protocol High performance data access Conventional file system semantics SKIP SKIP SKIP
- 8. Replication in practice Read-only replication Clumsy manual release model Lacks complex data sharing (concurrent writes) Optimistic replication Inconsistent consistency SKIP SKIP SKIP
- 9. Consistent replication Problem: state of the practice in file system replication does not satisfy the requirements of global scientific collaborations How can we provide Grid applications efficient and reliable data access? Consistent replication SKIP SKIP SKIP
- 10. Design principles Optimal read-only behavior Performance must be identical to un-replicated local system Concurrent write behavior Ordered writes, i.e., one-copy serializability Close-to-open semantics Fine-grained replication control The granularity of replication control is a single file or directory SKIP SKIP SKIP
- 11. Replication control client When a client opens a file for writing, the selected server temporarily becomes the primary for that file Other replication servers are instructed to forward client requests for that file to the primary if concurrent writes occur SKIP SKIP SKIP wopen
- 12. Replication control client The primary server asynchronously distributes updates to other servers during file modification SKIP SKIP SKIP write
- 13. Replication control client When the file is closed and all replication servers are synchronized, the primary server notifies the other replication servers that it is no longer the primary server for the file SKIP SKIP SKIP close
- 14. Directory updates Prohibit concurrent updates A replication server waits for the primary to relinquish its role Atomicity for updates that involve multiple objects (e.g. rename) A server must become primary for all objects Updates are grouped and processed together SKIP SKIP SKIP
- 15. Close-to-open semantics Server becomes primary after it collects votes from a majority of replication servers Use a majority consensus algorithm Cost is dominated by the median RTT from the primary server to other replication servers Primary server must ensure that every replication server has acknowledged its election when a written file is closed Guarantees close-to-open semantics Heuristic: allow a new file to inherit the primary server that controls its parent directory for file creation SKIP SKIP SKIP
- 16. Durability guarantee “ Active view” update policy Every server keeps track of the liveness of other servers (active view) Primary server removes from its active view any server that fails to respond to its request Primary server distributes updates synchronously and in parallel Primary server acknowledges a client write after a majority of replication servers reply Primary sends other servers its active view with file close A failed replication server must synchronize with the up-to-date copy before it can rejoin the active group I suppose this is expensive SKIP SKIP SKIP
- 17. What I skipped Not the Right Stuff GridFTP: manual synchronization NFSv4.r: write-mostly WAN performance AFS, Coda, et al.: sharing semantics Consistent replication for Grid computing Ordered writes too weak Strict consistency too strong Open-to-close just right
- 18. NFSv4.r in brief View-based replication control protocol Based on (provably correct) El-Abbadi, Skeen, and Cristian Dynamic election of primary server At the granularity of a single file or directory Majority consensus on open (for synchronization) Synchronous updates to a majority (for durability) Total consensus on close (for close-to-open)
- 19. Write-mostly WAN performance Durability overhead Synchronous updates Synchronization overhead Consensus management
- 20. Asynchronous updates Consensus requirement delays client updates Median RTT between the primary server and other replication servers is costly Synchronous write performance is worse Solution: asynchronous update Let application decide whether to wait for server recovery or regenerate the computation results OK for Grid computations that checkpoint Revisit at end with new ideas
- 21. Hierarchical replication control Synchronization is costly over WAN Hierarchical replication control Amortizes consensus management A primary server can assert control at different granularities
- 22. Shallow & deep control /usr bin local /usr bin local A server with a shallow control on a file or directory is the primary server for that single object A server with a deep control on a directory is the primary server for everything in the subtree rooted at that directory
- 23. Primary server election Allow deep control for a directory D if D has no descendent is controlled by another server Grant a shallow control request for object L from peer server P if L is not controlled by a server other than P Grant a deep control request for directory D from peer server P if D is not controlled by a server other than P No descendant of D is controlled by a server other than P SKIP SKIP SKIP
- 24. Ancestry table /root a b c f2 d2 controlled by S1 controlled by S0 controlled by S0 controlled by S2 …… Ancestry Table The data structure of entries in the ancestry table d1 f1 Ancestry Entry an ancestry entry has the following attributes id = unique identifier of the directory array of counters = set of counters recording which servers controls the directory’s descendants counter array S0 S1 S2 Id 2 0 0 c 0 0 1 b 2 1 0 a 2 1 1 root
- 25. Primary election S0 and S1 succeed in their primary server elections S2’s election fails due to conflicts Solution - S2 then re-tries by asking for shallow control of a a b c S0 S1 S2 control b control c control b deep control a control c deep control a S0 S1 S2 SKIP SKIP SKIP
- 26. Performance vs. concurrency Associate a timer with deep control Reset the timer with subsequent updates Release deep control when timer expires A small timer value captures bursty updates Issue a separate shallow control for a file written under a deep controlled directory Still process the write request immediately Subsequent writes on the file do not reset the timer of the deep controlled directory SKIP SKIP SKIP SKIP
- 27. Performance vs. concurrency Increase concurrency when the system consists of multiple writers Send a revoke request upon concurrent writes The primary server shortens releasing timer Optimally issues a deep control request for a directory that contains many updates in single writer cases SKIP SKIP SKIP
- 28. Single remote NFS N.B.: log scale
- 29. Deep vs. shallow Shallow controls vs. deep + shallow controls
- 31. Durability revisited Synchronization is expensive, but … When we abandon the durability guarantee, we risk losing the results of the computation And may be forced to rerun it But it might be worth it Goal: maximize utilization NEW NEW NEW
- 32. Utilization tradeoffs Adding synchronous replication servers enhances durability Which reduces the risk that results are lost And that the computation must be restarted Which benefits utilization But increases run time Which reduces utilization
- 33. Placement tradeoffs Nearby replication servers reduce the replication penalty Which benefits utilization Nearby replication servers are more vulnerable to correlated failure Which reduces utilization
- 34. Run-time model
- 35. Parameters F: failure free, single server run time C: replication overhead R: recovery time p fail : server failure p recover : successful recovery
- 36. F: run time Failure-free, single server run time Can be estimated or measured Our focus is on 1 to 10 days
- 37. C: replication overhead Penalty associated with replication to backup servers Proportional to RTT Ratio can be measured by running with a backup server a few msec away
- 38. R: recovery time Time to detect failure of the primary server and switch to a backup server We assume R << F Arbitrary realistic value: 10 minutes
- 39. Failure distributions Estimated by analyzing PlanetLab ping data 716 nodes, 349 sites, 25 countries All-pairs, 15 minute interval From January 2004 to June 2005 692 nodes were alive throughout We ascribe missing pings to node failure and network partition
- 41. Same-site correlated failures sites nodes 11 21 65 259 0.488 5 0.488 0.378 4 0.538 0.440 0.546 3 0.561 0.552 0.593 0.526 2
- 43. Run-time model Discrete event simulation yields expected run time E and utilization (F ÷ E)
- 44. Simulated utilization F = one hour One backup server Four backup servers
- 45. Simulation results F = one day One backup server Four backup servers
- 46. Simulation results F = ten days One backup server Four backup servers
- 47. Simulation results discussion For long-running jobs Replication improves utilization Distant servers improve utilization For short jobs Replication does not improve utilization In general, multiple backup servers don’t help much Implications for checkpoint interval …
- 48. Checkpoint interval F = one day One backup server 20% checkpoint overhead F = ten days, 2% checkpoint overhead One backup server Four backup servers
- 49. Next steps Checkpoint overhead? Replication overhead? Depends on amount of computation We measure < 10% for NAS Grid Benchmarks, which do no computation Refine model Account for other failures Because they are common Other model improvements
- 50. Conclusions Conventional wisdom holds that consistent mutable replication in large-scale distributed systems is too expensive to consider Our study proves otherwise
- 51. Conclusions Consistent replication in large-scale distributed storage systems is feasible and practical Superior performance Rigorous adherence to conventional file system semantics Improves cluster utilization
- 52. Thank you for your attention! www.citi.umich.edu Questions?