GTLAB: Grid Tag Libraries Supporting Workflows within Science Gateways Mehmet A. Nacar, Marlon E.
Download ReportTranscript GTLAB: Grid Tag Libraries Supporting Workflows within Science Gateways Mehmet A. Nacar, Marlon E.
GTLAB: Grid Tag Libraries Supporting Workflows within Science Gateways Mehmet A. Nacar, Marlon E. Pierce, Geoffrey C. Fox Community Grids Lab Indiana University Grid-style portal as used in Earthquake Grid The Portal is built from portlets – providing user interface fragments for each service that are composed into the full interface – uses OGCE technology as does planetary science VLAB portal with University of Minnesota Now to Portals 2 Introduction to GTLAB • Grid Tag Libraries and Beans (GTLAB) • Encapsulates clients to common Grid services as XML tag libraries and server side Java beans. – Embedded by portlet developers in their portlet pages to invoke common tasks – Specification of the composite action you want to occur when a user hits the submit button. – Allows portal developers to concentrate on the user interface components. • Tags can be arranged in directed acyclic graphs (dependency chains). – These represent simple workflows. • Based on extensions to Java Server Faces (JSF) and the Java CoG Kit. • Implements a component model for UI’s that is more powerful than portlets Motivation • OGCE Grid portlets typically wrap each single Grid capability in a separate portlet – GridFTP-->GridFTP Portlet • Gateway portlets encapsulate sophisticated but specialized functionality. – Such as submitting multiple linked jobs – Each functionality requires a separate portlet • We need a middle way to “automatically” build complex functionality from component parts – Traditional portlets don’t work as no agreed way to combine component portlets into an integrated user capability – We need a component model for portlets: reusable portlet parts • JSF is our starting point – Removes dependencies on the Servlet API. – Backend software is just beans, so can be reused more easily outside of web and portlet applications. • JSF also provides an extensible framework (tag libraries) • Apache JSF portlet bridge allows you to convert standalone JSF applications (development phase) into portlets (deployment phase). Architecture Overview • Grid portals are client to backend codes through Web/Grid services. • Grid tags are part of user interface tier and embedded into portlet container. • Grid tags use local services in Apache Tomcat to manage sessions and handlers. • Grid beans implement a layer on top of Grid client APIs such as Java CoG • Implies that “portal” is quite sophisticated as must support integration • Note CoG kit is here to act as a broker to hide complexities of evolving services (e.g. different versions of Globus) • “violates service model” in that core software centralized in portal application 5 GTLAB Example • Grid tags are associated with Grid services via Grid beans • Grid Beans wrap the Java COG Kit (version 4) • We show an example JSF page section below. • This allows you to develop new Grid portlets with no additional Java code. <html><body> < f:view> <!-- Other user interface tags go here--> <f:form> <o:submit id=”test” action=”next_page”> <o:myproxy id=”pr” hostname=”gf1.ucs.indiana.edu” port=”7512” lifetime=”2” username=“mnacar” password=”***” /> <o:jobsubmit id=”task” hostname=”cobalt.ncsa.teragrid.org” provider=”GT4” executable=”/bin/ls” stdout=”tmp/result stderr=”tmp/error” /> </o:submit> </f:form> </f:view> </body> </html> GTLAB features • GTLAB provides common components for building portlets using tags and reusable parts. • The goal of GTLAB to simplify Grid portlet development – Enable rapid development from reusable components • GTLAB capabilities include Grid operations with XML based tags within JSF framework. • Grid tag libraries are built using JSF custom component development techniques • Grid tags are interfaces to backing Grid beans – End users pass values to Grid beans by using tag attributes. • Each backing Grid bean has equal capability with a portlet application in case of Grid portlet approach. – GridFtp tag generates GridFTP Portlet • But tags allow service backend component model to map into an intermediate UI component models – each tag set is a component – The components tags can easily be composed into a rich UI which portlets cannot do 7 Grid Tags Associated Grid Beans Features <submit/> ComponentBuilderBean Creating components, job handlers, submitting jobs. This is visually rendered as HTML <handler/> MonitorBean Handling monitoring page actions <multitask/> MultitaskBean Constructing simple workflow <dependency/> MultitaskBean Defining dependencies among sub jobs <myproxy/> MyproxyBean Retrieving myproxy credential <fileoperation/> FileOperationBean Providing Gridftp operations <jobsubmission/> JobSubmitBean Providing GRAM job submissions <filetransfer/> FileTransferBean Providing Gridftp file transfer (Other JSF UI Tags) ResourceBean Describes common properties among all tags and beans. Passing values given by standard visual JSF components. VLAB Computational Chemistry Portal VLAB: The Virtual Laboratory for Earth and Planetary Materials • Primarily a traditional job submission, monitoring, and management portal. – Originally tried to use first generation of OGCE where all capabilities are portlets • Collaborative Grid services and portals support computational material science. • VLAB Challenges: – Generic GRAM job submission portlet does not fit with VLab requirements • Quantum Espresso requires parameter entries and checking – Grid Portlets must be easy to develop using component libraries. • • • • Transfer data files in and out of the desired remote host. Run one or more executables. Keep track of job progress Store all of the information as “job archive” for reproducibility. 10 Vlab portlet pages • VLab portlets include job preparation and submission, job monitoring • Vlab portlet pages can handle sophisticated user interfaces Case Study: VLAB Portal 12 Encoding DAGs in Portlets Multi-staged task Input Task A File Operation (mkdir) depends Task B File Transfer depends Task C Job Submission depends Task D File Transfer Output • Multitask provides a simple Directed Acyclic Graph (DAG) • This example demonstrates a composite Grid job using multistaged multitask • GTLAB handles lifecycle of DAG within JSF application DAG Example JSF Page <o:submit id=”test” action=”next_page” /> <o:multitask id=”mytask” taskname=”test” persistent=”true” > <o:myproxy id=”pr” hostname=”gf1.ucs.indiana.edu” port=”7512” lifetime=”2” username=“nacar” password=”***” /> <o:fileoperation id=”taskA” command=”mkdir” hostname=”cobalt.ncsa.teragrid.org” path=”/home/manacar/tmp/” /> <o:filetransfer id=”taskB” from=”gridftp://gf1.ucs.indiana.edu:2811/home/manacar/input_file” to=”gridftp://cobalt.ncsa.teragrid.org:2811/home/manacar/input_file” /> <o:jobsubmit id=”taskC” hostname=”cobalt.ncsa.teragrid.org” provider=”GT4” executable=”/bin/execute” stdin=”tmp/input_file” stdout=”tmp/result” stderr=”tmp/error” /> <o:filetransfer id=”taskD” from=”gridftp://cobalt.ncsa.teragrid.org:2811/home/manacar/tmp/result” to=” gridftp://gf1.ucs.indiana.edu:2811/home/manacar/result” /> <o:dependency id=”dep1” task=”taskB” dependsOn=”taskA” /> <o:dependency id=”dep2” task=”taskC” dependsOn=”taskB” /> <o:dependency id=”dep3” task=”taskD” dependsOn=”taskC” /> </o:multitask> </o:submit> This encodes the DAG on the previous page. • Grid tags are embedded into JSF view pages and decorated with standard JSF form, input, output and button components – Grid components are non-visual. GTLAB Standard JSF 15 DAG extensions: Condor DAGMan • GTLAB architecture is extensible – Besides Globus Grid services support, Grid community also use other common Grid services such as Condor • We extend GTLAB to support Condor capabilities – Condor DAGMan is a tool for complex application workflows on Condor – Birdbath is Web services provider of Condor capabilities – GTLAB tags for DAGMan use Birdbath service to create client stubs • Grid tags integrate DAGMan with the following tags: – <o:condorDagman/> and <o:condorSubmit/> • GTLAB executes and monitors DAGs by Condor • Composing DAGMan workflow is out of scope. 16 Taverna workflows • DAGs provide advantages to pipe Grid invocations (dataflow) • But DAGs cannot support full-fledged workflow capabilities like conditional branches and loops. – We studied Taverna as a test case. – We can investigate Kepler and BPEL implementations as future extensions • Workflow systems have different features: – Composing – Enacting – Monitoring • GTLAB supports Taverna enactment and monitoring. • GTLAB imports well studied built-in workflows collected by the community – Bioinformatics workflows and their metadata is available • Workflow composition is out of scope of this work – There is much ongoing research in this area 17 Taverna use case JSF Grid tags XML Scufl document load scufl document (3) (2) extract grid tags JSF Action HTTP submit execute (4) provide input parameters (4) load input parameters Taverna bean (5) execute Taverna enactor • A user interacts with a workflow portlet (left of picture) to utilize Taverna enactor. • User provides parameters by submitting a web form that start the sequence of events. (1) November 6, 2015 Mehmet Nacar 18 Advantages of GTLAB • GTLAB provides simplicity to develop science portals – Rapid development – Easy deployment • Grid tags provide rich selection of attributes to initialize Grid beans. • Composite tasks can contain an unlimited number of subtasks • GTLAB gives flexibility to developers to use their own Grid beans library or add more Grid beans to the existing ones. – Following the method name convention of GTLAB • Grid beans also can be imported to any presentation logic – GCEShell is a command-line tool • Grid bean methods are bound to tags with attributes to simplify the building of new Grid portlets GTLAB Overhead GTLAB Processing Time (msec) 2 JSF Processing 153 Handler storing 1 Submitting 410 • Total overhead = Tsubmit-Trequest = 156 msec (that includes JSF overhead) • Average overhead of GTLAB is about few milliseconds • GTLAB does not add up significant delay on processing the requests. 20 GTLAB Related Work • GridSphere’s Grid Portlets 1.3 – Grid Portlets 1.3 provide API and User Interface (UI) tags to build Grid portlets – An effort called Vine (Portlet Vine) refactors Grid portlets and decouples the portlets from GridSphere • Karajan – XML based workflow language and engine for Grid computing – Built on Java CoG Kit – Requires additional effort to aggregate it into Grid portals • MyGrid Portlet Interface (MPI) – Based on Taverna workflows (Scufl) – Workflow execution and monitoring portlets Conclusion and Future Work • Described the process for creating Grid portlets using GTLAB. – We gained rapid development by using reusable components • GTLAB can handle composite Grid operations • Extended our approach to support different workflow engines. – Condor DAGMan and Taverna • Investigate integrating AJAX and more generally gadget architectures (Web 2.0 portals) with our session management and monitoring. 22 More Information • GTLAB version 1.0 Beta release available at – http://grids.ucs.indiana.edu/users/manacar/GTLABwebsite • See link from main OGCE web site – http://www.collab-ogce.org • Contact OGCE: [email protected] • Contact Mehmet: [email protected] 23