Abstract
Quantum structure or scattering calculations often belong to a class of computational problems involving the aggregation of a set of matrices representing a linear problem to be solved. The authors discuss a number of approaches based on cluster and grid computing, and discuss the implementations and the respective merits and shortcomings. The authors consider MPI-based cluster computing in a self-scheduling paradigm, CoBRA (a cpu-harvesting desktop grid) in a farmer-worker paradigm, and a batch-computing paradigm on BEGrid (the Belgian research grid facility). It is observed that for all paradigms an efficient implementation is possible, yielding results within a comparable time frame.
TopBackground
Before focusing on the actual distribution, more insight into the actual problem and the possible distribution mechanisms is essential. This section discusses the actual nuclear physics problem and gives a brief introduction to each of the used distribution mechanisms (CoBRA, BeGrid and MPI)
Key Terms in this Chapter
Globus: The Globus Alliance is a community of organizations and individuals developing fundamental technologies behind the “Grid,” which lets people share computing power, databases, instruments, and other on-line tools securely across corporate, institutional, and geographic boundaries without sacrificing local autonomy. The Globus Toolkit is an open source software toolkit used for building Grid systems and applications. It is being developed by the Globus Alliance and many others all over the world. A growing number of projects and companies are using the Globus Toolkit to unlock the potential of grids for their cause.
BEgrid: Infrastructural Grid collaboration between several universities and research institutes in Belgium. BEgrid is part of the North European ROC region of the EGEE project.
Beowulf: Beowulf Clusters are scalable performance clusters based on commodity hardware, on a private system network, with open source software (Linux) infrastructure.
MPI: MPI stands for “Message Passing Interface”, and is a standard library specification for Message Oriented Middleware without modifications to compiler or system. It was designed for parallel and high performance computing
EGEE: EGEE stands for “Enabling Grids for E-sciencE”. THe project is funded by the European Commission and aims to build on recent advances in grid technology and develop a service grid infrastructure which is available to scientists 24 hours-a-day. It currently consists of some 41,000 CPU’s and some 5 PetaByte storage space, and typically runs about 100,000 jobs concurrently.
LCG: LCG stands for “Large Hadron Collider Computing Grid”, and its mission is to build and maintain data storage and analysis infrastructure for the entire high energy physics community that will use CERN’s large hadron collider. It contributed significantly to the development of computing grids, particularly in Europe.
MPICH-G2: MPICH-G2 is a grid-enabled implementation of the MPI v1.1 standard. Using services from the Globus Toolkit® (e.g., job startup, security), MPICH-G2 allows to couple multiple machines, potentially of different architectures, to run MPI applications. MPICH-G2 automatically converts data in messages sent between machines of different architectures and supports multiprotocol communication by automatically selecting TCP for intermachine messaging and (where available) vendor-supplied MPI for intramachine messaging
MPICH2: MPICH2 is an implementation of the Message-Passing Interface (MPI). The goals of MPICH2 are to provide an MPI implementation for important platforms, including clusters, SMPs, and massively parallel processors.