Cloud computing has seen tremendous growth, particularly for commercial web applications. The on-demand, pay-as-you-go model creates a flexible and cost-effective means to access compute resources. For these reasons, the scientific computing community has shown increasing interest in exploring cloud computing. However, the underlying implementation and performance of clouds are very different from those at traditional supercomputing centers. It is therefore critical to evaluate the performance of HPC applications in today's cloud environments to understand the tradeoffs inherent in migrating to the cloud. This work represents the most comprehensive evaluation to date comparing conventional HPC platforms to Amazon EC2, using real applications representative of the workload at a typical supercomputing center. Overall results indicate that EC2 is six times slower than a typical mid-range Linux cluster, and twenty times slower than a modern HPC system. The interconnect on the EC2 cloud platform severely limits performance and causes significant variability.

The combination of low-cost sensors, low-cost commodity computing, and the Internet is enabling a new era of data-intensive science. The dramatic increase in this data availability has created a new challenge for scientists: how to process the data. Scientists today are envisioning scientific computations on large scale data but are having difficulty designing software architectures to accommodate the large volume of the often heterogeneous and inconsistent data. In this paper, we introduce a particular instance of this challenge, and present our design and implementation of a MODIS satellite data reprojection and reduction pipeline in the Windows Azure cloud computing platform. This cloud-based pipeline is designed with a goal of hiding data complexities and subsequent data processing and transformation from end users. This pipeline is highly flexible and extensible to accommodate different science data processing tasks, and can be dynamically scaled to fulfill scientists' various computational requirements in a cost-efficient way. Experiments show that by running a practical large-scale science data processing job in the pipeline using 150 moderately-sized Azure virtual machine instances, we were able to produce analytical results in nearly 90X less time than was possible with a high-end desktop machine. To our knowledge, this is one of the first eScience applications to use the Windows Azure platform.

n e Grid2003 Project has deployed a multi-virfual organization, application-driven grid laboratory ('"Grid3'7 that has sustained for several months the production-level services required by physics experiments of the Large Hadron Collider at CERN (ATLAS and CMS), the Sloan Digital Sky Survey project, the gravitational wave search experiment LIGO, the BTeV mperiment at Fermilab, as well as applications in molecular structure analysis and genome analysis, and computer science research projects in such areas as job and data scheduling. The deployed infiastmcture has been operating since Noisniber 2003 with 27 sites, apeak of 2800 processors, work loads fiom 10 different applications exceeding 1300 simultaneous jobs, and data transfers among sites of greater than 2 TBiday. We describe the principles that have guided the development of this unique infrastructure and the practical experiences that have resultedfiom its creation and use. We discuss application requirements for grid services deployment and con$guration. monitoring infiastnic fure, application performance, metrics. and operational experiences. We also summarize lessons learned.

The requirement for calibrating transducers having subnanometre displacement sensitivities stimulated the development of an instrument in which the displacement is measured by a combination of optical and X-ray interferometry. The need to combine both types of interferometry arises from the fact that optical interferometry enables displacements corresponding to whole numbers of optical fringes to be measured very precisely, but subdivision of an optical fringe may give rise to errors that are significant at the subnanometre level. The X-ray interferometer is used to subdivide the optical fringes. Traceability to the meter is achieved via traceable calibrations of the lattice parameter of silicon and of the laser frequency. Polarization encoding and phase modulation allow the optical interferometer to be precisely set on a specific position of the interference fringe-the null point setting. The null point settings in the interference fringe field correspond to dark or bright hinges. Null measurement ensures maximum possible noise rejection. However, polarization encoding makes the interferometer nonlinear, but all nonlinearity effects are effectively zero at the fringe set point. The X-ray interferometer provides the means for linear subdivision of optical fringes. Each X-ray fringe corresponds to a displacement that is equal to the lattice parameter of silicon, which is ca. 0.19 nm for the (220) lattice planes. For displacements up to 1 mu m the measurement uncertainties at 95% confidence level are +/-30 pm, and for displacements up to 100 mu m and 1 mm the uncertainties are +/-35 and +/-170 pm, respectively. Important features of the instrument, which is located at the National Physical Laboratory, are the silicon monolith interferometer that both diffracts X-rays and forms part of the optical interferometer, a totally reflecting parabolic collimator for enhancing the usable X-ray flux and the servo-control for the interferometers

No abstract