PVM Grids to Self-assembling Virtual Machines
نویسنده
چکیده
Oak Ridge National Laboratory (ORNL) leads two of the five big Genomes-to-Life projects funded in the USA. As a part of these projects researchers at ORNL have been using PVM to build a computational biology grid that spans the USA. This talk will describe this effort, how it is built, and the unique features in PVM that led the researchers to choose PVM as their framework. The computations such as parallel BLAST are run on individual supercomputers or clusters within this P&P grid and are themselves written in PVM to exploit PVM’s fault tolerant capabilities. We will then describe our recent progress in building an even more adaptable distributed virtual machines package called Harness. The Harness project includes research on a scalable, selfadapting cote called H2O, and research on fault tolerant MPI. Harness software framework provides parallel software “plug-ins” that adapts the run-time system to changing application needs in real time. This past year, we have demonstrated Harness’ ability to self-assemble into a virtual machine specifically tailored for particular applications. Finally, we will describe DOE’s plan to create a National Leadership Computing Facility, which will house a 100 TF Cray X2 system, and a Cray Red Storm at ORNL, and an IBM Blue Gene system at Argonne National Laboratory (ANL). We will describe the scientific missions of this facility and the new concept of “computational end stations” being pioneered by the Facility. 1 Genomics Grid built on PVM The United States Department of Energy (DOE) has embarked on an ambitious computational biology program called Genomes-to-Life. The program is using DNA sequences from microbes and higher organisms, for systematically tackling questions about the molecular machines and regulatory pathways of living systems. Advanced technological and computational resources are being employed to identify and understand the underlying mechanisms that enable living organisms to develop and survive under a wide variety of environmental conditions. ORNL is a leader in two of the five Genomes-to-Life centers. As part of this effort ONRL is building a Genomics Computational Grid across the U.S. connecting ORNL, ANL, Pacific Northwest National Laboratory, and Lawrence Berkeley National Laboratory. The software being deployed is called The Genome Channel. The Genome Channel is a
منابع مشابه
Executing and Monitoring PVM Programs in Computational Grids with Jini
This paper presents a way to build a computational Grid for PVM programs. The Grid applies Jini to handle the dynamically changing set of participants, and to make the communication between Grid-clients and Grid-machines possible. In case of a PVM-Grid the grid-resources are Parallel Virtual Machines. Our system provides a high-level interface for them, which through the users can submit their ...
متن کاملProgress towards Petascale Virtual Machines
Petascale Virtual Machines (PVM) continues to be a popular software package both for creating personal grids and for building adaptable, fault tolerant applications. We will illustrate this by describing a computational biology environment built on top of PVM that is used by researchers around the world. We will then describe or recent progress in building an even more adaptable distributed vir...
متن کاملRunning Large-scale Applications on Cluster Grids 157 Running Large-scale Applications Running Large-scale Applications on Cluster Grids
Grid computing enables the enormous, but often poorly utilized, computing resources existing on the Internet to be harnessed in order to solve large-scale problems. However, if they want to solve the problems of configuration and performance optimization of their large-scale applications, programmers wishing to exploit such computing resources have to deal with highly variable communication del...
متن کاملSupporting Heterogeneous Network Computing: PVM
The Parallel Virtual Machine (PVM), an integrated framework for heterogeneous network computing, lets scientists exploit collections of networked machines when carrying out complex scientiic computations. Under PVM, a user-deened grouping of serial, parallel, and vector computers appears as one large distributed-memory machine. Connguring a personal parallel virtual computer involves simply lis...
متن کاملThe Performance of PVM on MPP Systems
PVM (Parallel Virtual Machine) is a popular standard for writing parallel programs so that they may execute over a network of heterogeneous machines. This paper presents some performance results of PVM on three massively parallel processing systems: the Thinking Machines CM-5, the Intel Paragon, and the IBM SP-2. We describe the basics of the communication model of PVM and its communication rou...
متن کامل