1. London e-Science Centre GridSAM A Standards Based Approach to Job Submission A. Stephen M C Gough Imperial College London A Standards Based Approach to Job Submission A. Stephen M C Gough Imperial College London

2. London e-Science Centre 2 Common Grid Systems There are many Grid Systems.  Though they are not compatible.  Globus  Condor  gLite  Grid Engine  LSF  PBS  LoadLeveler  Unicore  … There are many Grid Systems.  Though they are not compatible.  Globus  Condor  gLite  Grid Engine  LSF  PBS  LoadLeveler  Unicore  … Different job description languages Different submission interfaces

3. London e-Science Centre 3 If all DRM systems supported the same interface…  If we had:  One interface definition for job submission  One job description language  Then life would be easier!  We’re getting there  JSDL is a proposed standard job submission description language  OGSA-BES is a proposed standard basic execution service interface  One day hopefully everyone will support this  Till then…  If we had:  One interface definition for job submission  One job description language  Then life would be easier!  We’re getting there  JSDL is a proposed standard job submission description language  OGSA-BES is a proposed standard basic execution service interface  One day hopefully everyone will support this  Till then…

4. London e-Science Centre GridSAM Job Submission and Monitoring Web Service

5. London e-Science Centre 5 GridSAM Overview Grid Job Submission and Monitoring Service  What is GridSAM?  A Job Submission and Monitoring Web Service  Funded by the Open Middleware Infrastructure Institute (OMII) managed programme  V2.0.0 Available as part of the OMII 2.x release  Open source (BSD)  One of the first system to support the GGF Job Submission Description Language (JSDL)  What is GridSAM?  A Job Submission and Monitoring Web Service  Funded by the Open Middleware Infrastructure Institute (OMII) managed programme  V2.0.0 Available as part of the OMII 2.x release  Open source (BSD)  One of the first system to support the GGF Job Submission Description Language (JSDL)

6. London e-Science Centre 6 What’s not?  GridSAM is not  a scheduling service  That’s the role of the underlying launching mechanism  That’s the role of a super-scheduler that brokers jobs to a set of GridSAM services  a provisioning service  GridSAM runs what’s been told to run  GridSAM does not resolve software dependencies and resource requirements  GridSAM is not  a scheduling service  That’s the role of the underlying launching mechanism  That’s the role of a super-scheduler that brokers jobs to a set of GridSAM services  a provisioning service  GridSAM runs what’s been told to run  GridSAM does not resolve software dependencies and resource requirements

7. London e-Science Centre 7 GridSAM Overview Grid Job Submission and Monitoring Service  What is GridSAM to the resource owners?  A Web Service to expose heterogeneous execution resources uniformly  Single machine through Forking or SSH  Condor Pool  Grid Engine 6 through DRMAA  Globus 2.4.3 exposed resources  OR use our plug-in API to implement …  What is GridSAM to the resource owners?  A Web Service to expose heterogeneous execution resources uniformly  Single machine through Forking or SSH  Condor Pool  Grid Engine 6 through DRMAA  Globus 2.4.3 exposed resources  OR use our plug-in API to implement …

8. London e-Science Centre 8 GridSAM Overview Grid Job Submission and Monitoring Service  What is GridSAM to end-users?  A set of end-user tools and client-side APIs to interact with a GridSAM web service  Submit and Start Jobs  Monitor Jobs  Terminate Jobs  File transfer  Client-side submission scripting  Client-side Java API  What is GridSAM to end-users?  A set of end-user tools and client-side APIs to interact with a GridSAM web service  Submit and Start Jobs  Monitor Jobs  Terminate Jobs  File transfer  Client-side submission scripting  Client-side Java API

9. London e-Science Centre 9 Integration with OMII Distribution

10. London e-Science Centre 10 GridSAM Implementation  Virtual File System API (Apache VFS)  FTP / GSIFTP / HTTP / WEBDAV / SFTP  POSIX Shell API  Fork / SSH  Event dispatches (OpenSymphony Quartz)  Job Persistence (Hibernate - JDBC databases)  Runtime Monitoring and Control (Java Management Extension)  Virtual File System API (Apache VFS)  FTP / GSIFTP / HTTP / WEBDAV / SFTP  POSIX Shell API  Fork / SSH  Event dispatches (OpenSymphony Quartz)  Job Persistence (Hibernate - JDBC databases)  Runtime Monitoring and Control (Java Management Extension)

11. London e-Science Centre 11 Example Pipeline: Condor GridSAM Architecture  A staged event-driven architecture*  Submission pipeline is constructed as a network of stages connected by event queues  Each stage perform specific action upon incoming events  A staged event-driven architecture*  Submission pipeline is constructed as a network of stages connected by event queues  Each stage perform specific action upon incoming events *M. Welsh and D. Culler and E. Brewer. Seda: An architecture for well- connected scalable internet services. In Eighteenth Symposium on Operating Systems Principles (SOSP-18), October 2001.

12. London e-Science Centre 12 Deployment Scenario: Forking HTTP + WS-Sec./ HTTPS + WS- Sec. / HTTPS mutual. Local FS Local FS GSIFTP FTP WEBDAV HTTP …

13. London e-Science Centre 13 Deployment Scenario: Secure Shell (SSH) HTTP + WS-Sec./ HTTPS + WS- Sec. / HTTPS mutual. GSIFTP FTP WEBDAV HTTP … SFTP - FS SFTP - FS

14. London e-Science Centre 14 Deployment Scenario: Condor Pool Condor command- line wrapper HTTP + WS-Sec./ HTTPS + WS-Sec. / HTTPS mutual. GSIFTP FTP WEBDAV HTTP … Network FS Network FS

15. London e-Science Centre 15 Deployment Scenario: Globus 2.4.3

16. London e-Science Centre 16 Deployment Scenario: Grid Engine 6 GSIFTP FTP WEBDAV HTTP … Network FS Network FS

17. London e-Science Centre 17 Latest Features  Available in v2.0.0 (released 1/7/06)  MPI Application through GT2 plugin  Simple non-standard JSDL extension that extends with a element  Authorisation based on JSDL structure  Allow / deny submission based on a set of XPath rules and the identities of the submitter (e.g. distinguished name).  Prototype Basic Execution Service (ogsa-bes) interface  Demonstrated in the mini face-to-face in London last December  Shown interoperability with the Uni. Of Virginia BES (.NET based) implementation.  Available in v2.0.0 (released 1/7/06)  MPI Application through GT2 plugin  Simple non-standard JSDL extension that extends with a element  Authorisation based on JSDL structure  Allow / deny submission based on a set of XPath rules and the identities of the submitter (e.g. distinguished name).  Prototype Basic Execution Service (ogsa-bes) interface  Demonstrated in the mini face-to-face in London last December  Shown interoperability with the Uni. Of Virginia BES (.NET based) implementation.

18. London e-Science Centre 18 Upcoming Features  New DRMConnectors  PBS, EGEE, LSF, CCS  GridBS  Using the Condor matchmaker with GridSAM  With Oxford University  Basic Execution Service  Continue tracking the changes in the ogsa- bes specification  Support dual submission WS-interfaces  New DRMConnectors  PBS, EGEE, LSF, CCS  GridBS  Using the Condor matchmaker with GridSAM  With Oxford University  Basic Execution Service  Continue tracking the changes in the ogsa- bes specification  Support dual submission WS-interfaces

19. London e-Science Centre 19 GridSAM deployment  UK NGS  Supported by the Belfast e-Science Centre  UCL  Newcastle  Reading University  Imperial College London  …  JSDL Portlet from DL  UK NGS  Supported by the Belfast e-Science Centre  UCL  Newcastle  Reading University  Imperial College London  …  JSDL Portlet from DL

20. London e-Science Centre 20 GridSAM and AHE  The Application Hosting Environment (AHE) uses GridSAM for deploying their application wrappers to resources on the Grid.  As part of the SPICE (Simulated Pore Interactive Computing Environment) demonstrator for SuperComputing’05 GridSAM (along with the AHE) was used to deploy exeutables on resources both part of the UK National Grid Service and the US TeraGrid.  This helped towards SPICE winning the HPC Analytics Challenge for demonstrating the use of innovative techniques in rigorous data analysis and high-end visualisation to solve a complex, real-world problem.  The Application Hosting Environment (AHE) uses GridSAM for deploying their application wrappers to resources on the Grid.  As part of the SPICE (Simulated Pore Interactive Computing Environment) demonstrator for SuperComputing’05 GridSAM (along with the AHE) was used to deploy exeutables on resources both part of the UK National Grid Service and the US TeraGrid.  This helped towards SPICE winning the HPC Analytics Challenge for demonstrating the use of innovative techniques in rigorous data analysis and high-end visualisation to solve a complex, real-world problem.

21. London e-Science Centre 21 GridSAM and e-protein  e-protein is performing protein annotation  Using GridSAM for executing bioinformatics jobs on the Grid  e-protein is performing protein annotation  Using GridSAM for executing bioinformatics jobs on the Grid Web Interface DRM Job Scheduler Java Control Program Workflow Pipeline Job 3D- GENOMICS Web Portal DRM Job Control Program Workflow GridSAM Job Job Description document Job Identifier Sequence files SSH tunnel Dasty Viewer Level 1 Level 2 Level 3

22. London e-Science Centre 22 GridSAM and MicroArray  Simple, robust access to up-to-date resources for microarray data storage and analysis  Distributed availability of the portal, allowing access to large compute power facilities required for microarray analysis and storage  Simple, robust access to up-to-date resources for microarray data storage and analysis  Distributed availability of the portal, allowing access to large compute power facilities required for microarray analysis and storage Grid Compute cluster EMAAS Server Scatter Plots Simple Affy PLM plots

23. London e-Science Centre 23 GridSAM and eMinerals

24. London e-Science Centre 24 GridSAM at UCL GridSAM SGE Fork Condor Client Middle-tier JSDL SOAP Integrating GridSAM with BPEL Providing Condor WS DRMConnector Condor Submission machine(s) Schedd SOAP ClassAd Condor-WS

25. London e-Science Centre 25 GridSAM and NeuroGrid The aim of the NeuroGrid consortium is to enhance collaboration between clinical, imaging and e-scientists to create a Grid-based network of neuroimaging centres and a neuroimaging tool-kit. Sharing data, experience and expertise will facilitate the archiving, curation, retrieval and analysis of imaging data from multiple sites and enable large-scale clinical studies.

26. London e-Science Centre 26 GridSAM and HPCP Interop  GridSAM was one of over 10 OGSA- BES (HPCP profile) services demonstrating interop  Managed to interop with all but one other system  Profile was restrictive  Simple Security  No data staging  Taking part in the SC07 interop  GridSAM was one of over 10 OGSA- BES (HPCP profile) services demonstrating interop  Managed to interop with all but one other system  Profile was restrictive  Simple Security  No data staging  Taking part in the SC07 interop

27. London e-Science Centre Questions? Official Download http://www.omii.ac.uk Project Information and Documentation http://gridsam.sourceforge.net Official Download http://www.omii.ac.uk Project Information and Documentation http://gridsam.sourceforge.net