2009 IWPLS’09 Pages 1–8 G-FLUXO: A workflow portal specialized in Computational BioChemistry E. Gutierrez´ 1,∗, A. Costantini 2, J. Lopez´ Cacheiro 1,∗ and A. Rodr´ıguez 1∗ 1CESGA. Santiago de Compostela. Spain. 2Department of Chemistry, University of Perugia. Perugia, Italy. Associate Editor: Sandra Gesing and Jano van Hemert ABSTRACT Accelrys Materials Studio (http://accelrys.com/products/materials- The development of a Grid Portal Workflow aware and specialized studio) are commercial examples where an integration of in Computational BioChemistry is presented. specific scientific applications, execution on remote resources The P-GRADE Portal (GridSphere based) has been expanded with and centralized visualization have been implemented. However, specific portlet developments adding both support for Computational although a lot of work about applications integration (compatibility, BioChemistry applications and for different distributed resources is input/output format interchange, visualization...) have been done not described. too much effort have been done with the aim of an easy and efficient As a first prototype, GROMACS, a versatile package to perform execution of applications on the very different computational Molecular Dynamics simulations making use of the Newtonian platforms available today (from a multicore desktop to the equations of motion for systems with hundreds to millions of particles, heterogeneous and geographically distributed Grid infrastructure has been implemented and supported. Starting from that, specific passing through the cluster platform located on a lab or on a DAG workflows running GROMACS jobs, demanding very different supercomputer center). In this sense the G-Fluxo project is devoted computational resources on different local and Grid infrastructures to the development of a Grid Portal Workflow specialized in (i.e. EGEE, EELA, etc), have been developed and tested. The JMOL Computational BioChemistry where very different computational based portlet tighly integrated into the portal has been developed in platforms can be used without the need of very specific computer order to help the user in the visualization of both workflow progress skills. G-Fluxo is developed trying to use existing and widely and results. used technology in order to avoid very specific requirements for adding computational resources. The P-GRADE Portal (Kacsuk and Sipos (2005)) have been chosen as the starting point of this project so the most widely used Grid middleware is supported. 1 INTRODUCTION Integration of the cluster platform is based on the SSH protocol In the XX century advances in BioChemistry and Computer Science and the DRMAA standard (Tr¨oger et al. (2007)) operative on enabled the development of new mathematical models able to most of the clusters running today. Finally a GridSphere based simulate the behavior of complex systems at molecular level, solution (http://www.gridsphere.org) such as P-GRADE Portal also allowing to understand chemical reactions as well as macroscopic lets the development of applications specific portlets fully JSR 168 properties of such systems. The huge number of atoms and compliant (http://www.jcp.org/en/jsr/detail?id=168). interactions involved in the calculations require an increasing The first application chosen to be implemented and supported into amount of computational resources not always available in local the G-Fluxo project is GROMACS (van der Spoel et al. (2005)). laboratories. GROMACS is a suite of applications for the simulation of complex Advances in Computational Science as well as improvements systems making use of a wide variety of Molecular Dynamics in CPU performance, new architecture designs and the increasing techniques optimized for different architectures. The variety of availability of computer power on Grid platforms, together with simulations needed to solve a particular problem makes GROMACS the adoption of new computational models and algorithms adapted suitable to exploit different types of computer platforms to solve to better exploit parallelism and memory management of such one single problem. Workflows where each part can be run on very resources, gave a strong contribution to the computing simulation different platforms is possible, requiring to send one part of those of complex systems. jobs to High Performance Computing (HPC) and the other part to Nowadays many Graphical User Interfaces for BioChemistry High Throughput Computing (HTC). programs have been developed. Special emphasis have been Computing Grid resources can be considered as a type of put on the coordinate and collaborative use of different HTC platform, on the other hand, local clusters with high speed applications in order to perform more complex simulations. interconnect networks are examples of HPC platforms. In the Schr¨odinger Software (http://www.schrodinger.com), Scienomics present paper a system to split workflows between grid and MAPS (http://www.scienomics.com/Products/maps/index.php) and local clusters in a transparent way for the user is described. Traditionally the management of such workflows is performed manually, increasing notably the effort for the scientist to take ∗g-fl[email protected] 1 E. Gutierrez´ et al advantage of both platforms. The use of the developed web portal OGCE GridPort Vine P- provides an abstraction of these systems showing a user friendly, GRADE and customizable, interface. Portal The Web Portal presented in this article makes use of both HPC Framework GridSphere GridSphere GridSphere GridSphere and HTC infrastructures showing, as an example, an interface for 2.1.5 2.0.2 3.1 2.2.10 job submission using GROMACS. GROMACS output visualization is done through a specific developed portlet based on Jmol GLite no no yes yes (http://www.jmol.org). support The GROMACS and Jmol packages have been implemented also in the COMPCHEM P-GRADE Portal (http://ui.grid.unipg.it:8080/ Workflow yes (through no no yes (through gridsphere/gridsphere) web portal in order to test the porting of the editor Java Web Start) Java Web Start) portlet applications developed at CESGA. The paper is organized as follows: in Section 2 the design Web portlets portlets portlets portlets and implementation of the portal built upon existing technologies interface and Adobe is illustrated (a comparison of existing portals is also included); Flex in Section 3 a case study consisting of a workflow based on GROMACS is presented. Our conclusions and future work are no or no or no or summarized in section 4. VO easy (but certificate difficult difficult difficult insecure) management 2 PORTAL DESIGN AND IMPLEMENTATION Table 1. Comparison of existing Portal technologies Nowadays the most common way of sending jobs is through the command line, both for grid and local cluster. An alternative to sending jobs through command line is to use a graphical interface. A web portal has been chosen as the graphical interface, because it P-GRADE Portal was the portal chosen as the base for our work offers wide compatibility, between different platforms, on the client because it is Open Source, it supports gLite 3.1, it has a graphical side. Furthermore each user can access to their data from different editor for workflows, and it is able to manage the certificates that places and machines, and they don’t need to be worried about new each user has for each Virtual Organization (VO) of the grid. software instalation or updates. In the following subsections, some existing technologies to send 2.1.2 Portal Technology Servlets are implementations of Java simulations through a web portal will be described. Then, with an code that run at the server-side as an answer to a web client request. existing web portal as a base, the modifications in that portal will The result of that execution is a web page. The concept of ”portlet” be described, both to send jobs to local cluster and to visualize is closely related to servlet concept. The main different is that while simulation job output. a servlet generates a whole web page a portlet generates only one part of a given page. Portlets are defined in the Sun specification 2.1 Existing Technologies JSR 168. A study of existing technologies has been done where four existing From the user’s point of view, a portlet is often represented web portals were compared, the results are presented in this Section. visually by a window embedded in a web page, with icons to get Afterwards, the characteristics of the portal chosen as the starting help, to maximize,. That is called “modes” and can be specified point for our developments will be commented with more detail. in a XML configuration file. Portlets can be grouped into tabs, this approach allows the creation of separated sections in a web page, 2.1.1 Portal Comparison There are several implementations providing more versatility in composition, to the administrator of a of web portals for sending jobs to grid. Four Open Source site and even to a user. web portals were tested to use as a base for our work: Vine Portlets are usually written in Java and JSP, with some XML (http://vinetoolkit.org), GridPort 4.0.1 (http://gridport.net/main), configuration files, and deployed, in a portlet container at the server, OGCE Release 2 (http://www.ogce.org), and P-GRADE Portal. All using a script language as Jackarta Ant. of these portals have a common execution platform: GridSphere. P-GRADE Portal is implemented using servlets and portlets. In As it is shown in the Table 2.1.1 the Vine is remarkable for its fact, P-GRADE Portal can be considered as a set of portlets which interface based on Adobe Flex (http://www.adobe.com/go/flex); for are distributed with its dependencies. Those portlets are mainly example, it has a comfortable file manager for grid. Additionally it focused on grid computing. There are portlets to send jobs and has an API for development. With that API it is possible to develop to manage certificates (using MyProxy), files in Store Elements applications for the web portal and also for applications running in (SE), and user accounts.