The Ibis Project: Simplifying Grid Programming & Deployment Henri Bal bal@cs.vu.nl Vrije Universiteit Amsterdam

1. The Ibis Project:Simplifying Grid Programming & DeploymentHenri Balbal@cs.vu.nlVrije Universiteit Amsterdam

3. The ‘Promise of the Grid’ Efficient and transparent (i.e. easy-to-use) wall-socket computing over a distributed set of resources [Sunderam ICCS’2004, based on Foster/Kesselman]

4. Parallel computing on grids • Mostly limited to • trivially parallel applications • parameter sweeps, master/worker • applications that run on one cluster at a time • use grid to schedule application on a suitable cluster • Our goal: run real parallel applications on a large-scale grid, using co-allocated resources

5. Efficient wide-area algorithms • Latency-tolerant algorithms with asynchronous communication • Search algorithms (Awari-solver [CCGrid’08]) • Model checkers (DiVinE [PDMC’08]) • Algorithms with hierarchical communication • Divide-and-conquer • Broadcast trees • …..

6. Reality: ‘Problems of the Grid’ • Performance & scalability • Heterogeneous • Low-level & changingprogramming interfaces • writing & deploying grid applications is hard • Connectivity issues • Fault tolerance • Malleability ! User Wide-Area Grid Systems

7. The Ibis Project • Goal: • drastically simplify grid programming/deployment • write and go!

8. Approach (1) • Write & go: minimal assumptions about execution environment • Virtual Machines (Java) deal with heterogeneity • Use middleware-independent APIs • Mapped automatically onto middleware • Different programming abstractions • Low-level message passing • High-level divide-and-conquer

9. Approach (2) • Designed to run in dynamic/hostile grid environment • Handle fault-tolerance and malleability • Solve connectivity problems automatically (SmartSockets) • Modular and flexible: can replace Ibis components by external ones • Scheduling: Zorilla P2P system or external broker

10. Global picture

11. Applications Satin: divide & conquer Communication layer (IPL) SmartSockets Zorilla P2P JavaGAT Rest of talk

12. Outline • Grid programming • IPL • Satin • SmartSockets • Grid deployment • JavaGAT • Zorilla • Applications and experiments

13. Ibis Design

14. Ibis Portability Layer(IPL) • Java-centric “run-anywhere” library • Sent along with the application (jar-files) • Point-to-point, multicast, streaming, …. • Efficient communication • Configured at startup, based on capabilities (multicast, ordering, reliability, callbacks) • Bytecode rewriter avoids serialization overhead

15. Serialization • Based on bytecode-rewriting • Adds (de)serialization code to serializable types • Prevents reflection overheadduring runtime JVM Javacompiler bytecoderewriter source bytecode bytecode JVM JVM

16. Membership Model • JEL (Join-Elect-Leave) model • Simple model for tracking resources, supports malleability & fault-tolerance • Notifications of nodes joining or leaving • Elections • Supports all common programming models • Centralized and distributed implementations • Broadcast trees, gossiping

17. Programming models • Remote Method Invocation (RMI) • Group Method Invocation (GMI) • MPJ (MPI Java 'standard') • Satin (Divide & Conquer)

18. Satin: divide-and-conquer • Divide-and-conquer isinherently hierarchical • More general thanmaster/worker • Cilk-like primitives (spawn/sync) in Java • Supports malleability and fault-tolerance • Supports data-sharing between different branches through Shared Objects

19. Satin implementation • Load-balancing is done automatically • Cluster-aware Random Stealing (CRS) • Combines Cilk’s Random Stealing with asynchronous wide-area steals • Self-adaptive malleability and fault-tolerance • Add/remove machines on the fly • Survive crashes by efficientrecomputations/checkpointing

20. Self-adaptation with Satin • Adapt #CPUs to level of parallelism • Migrate work from overloaded to idle CPUs • Remove CPUs with poor network connectivity • Add CPUs dynamically when • Level of parallelism increases • CPUs were removed or crashed • Can also remove/add entire clusters • E.g., for network problems [Wrzesinska et al., PPoPP’07 ]

21. Approach • Weighted Average Efficiency (WAE): 1/#CPUs * Σspeedi * (1 – overheadi ) overheadis fraction idle+communication time speedi= relative speed of CPUi (measured periodically) • General idea: Keep WAE between Emin (30%) and Emax(50%)

22. Overloaded network link Iteration duration Iteration • Uplink of 1 cluster reduced to 100 KB/s • Remove badly connected cluster, get new one

23. Connectivity Problems • Firewalls & Network Address Translation (NAT) restrict incoming traffic • Addressing problems • Machines with >1 network interface (IP address) • Machine on a private network (e.g., NAT) • No direct communication allowed • E.g., between compute nodes and external world

24. SmartSockets library • Detects connectivity problems • Tries to solve them automatically • With as little help from the user as possible • Integrates existing and several new solutions • Reverse connection setup, STUN, TCP splicing, SSH tunneling, smart addressing, etc. • Uses network of hubs as a side channel

25. Example

26. Example [Maassen et al., HPDC’07 ]

27. Zorilla P2P JavaGAT Overview

28. JavaGAT • GAT: Grid Application Toolkit • Makes grid applications independent of the underlying grid infrastructure • Used by applications to access grid services • File copying, resource discovery, job submission & monitoring, user authentication • API is currently standardized (SAGA) • SAGA implemented on JavaGAT

29. Grid Applications with GAT Grid Application File.copy(...)‏ submitJob(...)‏ GAT Remote Files Monitoring Info service Resource Management GAT Engine GridLab Globus Unicore SSH P2P Local Intelligentdispatching globus gridftp [van Nieuwpoort et al., SC’07 ]

30. Zorilla: Java P2P supercomputing middleware

31. Zorilla components • Job management • Handling malleability and crashes • Robust Random Gossiping • Periodic information exchange between nodes • Robust against Firewalls, NATs, failing nodes • Clustering: nearest neighbor • Flood scheduling • Incrementally search for resources at more and more distant nodes [Drost et al., HPDC’07 ]

32. Overview

33. Ibis applications • e-Science (VL-e) • Brain MEG-imaging • Mass spectroscopy • Multimedia content analysis • Various parallel applications • SAT-solver, N-body, grammar learning, … • Other programming systems • Workflow engine for astronomy (D-grid), grid file system, ProActive, Jylab, …

34. Overview experiments • DAS-3: Dutch Computer Science grid • Satin applications on DAS-3 • Zorilla desktop grid experiment • Multimedia content analysis • High resolution video processing

35. DAS-3 272 nodes(AMD Opterons) 792 cores 1TB memory LAN: Myrinet 10G Gigabit Ethernet WAN (StarPlane): 20-40 Gb/s OPN Heterogeneous: 2.2-2.6 GHz Single/dual-core Delft no Myrinet

36. Gene sequence comparison in Satin (on DAS-3) Speedup on 1 cluster Run times on 5 clusters • Divide&conquer scales much better than master-worker • 78% efficiency on 5 clusters (with 1462 WAN-msgs/sec)

37. Barnes-Hut (Satin) on DAS-3 Speedup on 1 cluster Run times on 5 clusters • Shared object extension to D&C model improves scalability • 57% efficiency on 5 clusters (with 1371 WAN-msgs/sec)

38. Zorilla Desktop Grid Experiment • Small experimental desktop grid setup • Student PCs running Zorilla overnight • PCs with 1 CPU, 1GB memory, 1Gb/s Ethernet • Experiment: gene sequence application • 16 cores of DAS-3 with Globus • 16 core desktop grid with Zorilla • Combination, using Ibis-Deploy

39. 877 sec 3574 sec 1099 sec Ibis-Deploy deployment tool • Easy deployment with Zorilla, JavaGAT & Ibis-Deploy

40. Multimedia content analysis • Analyzes video streams to recognize objects • Extract feature vectors from images • Describe properties (color, shape) • Data-parallel task implemented with C++/MPI • Compute on consecutive images • Task-parallelism on a grid

41. MMCA application Ibis (Java) Client (Java) Parallel Horus Server Parallel Horus Servers Servers (C++) (local desk-top machine) Broker (Java) (grid) (any machine world-wide)

42. MMCA with Ibis • Initial implementation with TCP was unstable • Ibis simplifies communication, fault tolerance • SmartSockets solves connectivity problems • Clickable deployment interface • Demonstrated at manyconferences (SC’07) • 20 clusters on 3 continents, 500-800 cores • Frame rate increased from 1/30 to 15 frames/sec [Seinstra et al., IEEE Multimedia’07 ]

43. ‘Most Visionary Research’ award at AAAI 2007, (Frank Seinstra et al.) MMCA

44. High Resolution Video Processing Realtime processing of CineGrid movie data 3840x2160 (4xHD) @ 30 fps = 1424 MB/sec Multi-cluster processing pipeline Using DAS-3, StarPlane and Ibis

45. CineGrid with Ibis Use of StarPlane requires no configuration StarPlane is connected to local Myrinet network Detected & used automatically by SmartSockets Easy setup of application pipeline Connection administration of application is simplified by the IPL election mechanism Simple multi-cluster deployment (Ibis-Deploy) Uses Ibis serialization for high throughput

46. Summary • Goal: Simplify grid programming/deployment • Key ideas in Ibis • Virtual machines (JVM) deal with heterogeneity • High-level programming abstractions (Satin) • Handle fault-tolerance, malleability, connectivity problems automatically (Satin, SmartSockets) • Middleware-independent APIs (JavaGAT) • Modular

47. Acknowledgements Past members John Romein Gosia Wrzesinska • Rutger Hofman • Maik Nijhuis • Olivier Aumage • Fabrice Huet • Alexandre Denis • Current members • Rob van Nieuwpoort • Jason Maassen • Thilo Kielmann • Frank Seinstra • Niels Drost • Ceriel Jacobs • Kees Verstoep • Roelof Kemp • Kees van Reeuwijk

48. More information • Ibis can be downloaded from • http://www.cs.vu.nl/ibis • Papers: • Satin [PPoPP’07], SmartSockets [HPDC’07], Gossiping [HPDC’07], JavaGAT [SC’07],MMCA [IEEE Multimedia’07] • Ibis tutorials • Next one at CCGrid 2008 (19 May, Lyon)