The HPC Testbed of the Italian Grid Infrastructure

Abstract

Even though the Italian Grid Infrastructure (IGI) is a general purpose distributed platform, in the past it has been used mainly for serial computations. Parallel applications have been typically executed on supercomputer facilities or, in case of "not high-end" HPC applications, on local commodity parallel clusters. Nowadays, with the availability of multiple cores processors, Grid computing is becoming very attractive also for parallel applications but some problems exist in supporting of HPC applications on Grid environment. Here we describe the work made to set up a HPC testbed for "not high-end" HPC applications, based on IGI Grid technologies, to find solutions to those problems. Participating sites have been selected among the ones running HPC clusters in Grid environment. Each of them contributed with their specific HPC experience and their available resources to the present test, which encompasses an unprecedented large set of applications from different disciplines in the fields of astronomy, astrophysics, chemistry, climatology, material science and oceanography. In addition to computing resources sharing, the main contribution of each participant was the identification of the real requirements of his application also related to the current middleware limitations and then the realization of a test platform enhanced with additional HPC solutions and configurations developed in a tight collaboration between HPC administrators, users and IGI managers. The main work was on computational resources selection, data management and the definition, the deployment and the documentation of the software execution environment. The outcoming results of the testbed represent the basis of the HPC support in the IGI production infrastructure.

Porting workflows based on small and medium parallelism applications to the Italian Grid Infrastructure

“…On the contrary if the site is not partitioned it is important to set local policies to guarantee that the resources needed by parallel jobs are made available in a reasonable time. Solutions to this kind of problems have been discussed in [5].…”

“…Analyzing the use cases of the Italian communities belonging to such disciplines, in particular from Computational Chemistry, Earth Science and Bioinformatics (all of them were interested in exploiting the IGI computing capabilities) it was evident that better support of parallel applications was needed. In mid 2011 a working group has been set up to analyze the issues connected with parallel applications on Grid and to improve their support within the IGI infrastructure [5]. The working group took advantage of the new capabilities of the gLite middleware in supporting MPI and parallel jobs ( §3.1) and of the effort that the EGI project put in the same topic through a dedicated MPI Virtual Team [3].…”

Cloud Infrastructures forIn SilicoDrug Discovery: Economic and Practical Aspects

“…With respect to our previous works, we improved our brokering system in order to be able to interact with it. Although this tool is installed on a few INFN grid sites [51], which limits the size of the simulation that is possible to perform, it has a great potential. As regards the aspects related to the QoS requirements, they depend on the resource provider; therefore, for high-level request it is necessary to exploit commercial services different from the grid initiatives.…”