Scientific Computing in the Cloud

Univ. of Washington, Seattle, WA, USA
Computing in Science and Engineering (Impact Factor: 1.73). 07/2010; DOI: 10.1109/MCSE.2010.70
Source: IEEE Xplore

ABSTRACT Large, virtualized pools of computational resources raise the possibility of a new, advantageous computing paradigm for scientific research. To help achieve this, new tools make the cloud platform behave virtually like a local homogeneous computer cluster, giving users access to high-performance clusters without requiring them to purchase or maintain sophisticated hardware.

  • [Show abstract] [Hide abstract]
    ABSTRACT: Parallelization of time-dependent partial differential equations (PDEs) can be accomplished by time decomposition using the parareal algorithm. While the parareal algorithm was designed to enable real-time simulations, it holds particular promise for long time simulations on computational grids and clouds, due its low communication overhead and potential for adaptation to heterogeneous processors. This contribution extends previous work on the scheduling of tasks of the parareal algorithm to resources with heterogeneous CPU performance. Experiments on Amazon's EC2 show the suitability of this algorithm for execution on a heterogeneous cloud platform and its insensitivity to network latency.
    Parallel and Distributed Processing Symposium Workshops & PhD Forum (IPDPSW), 2012 IEEE 26th International; 01/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: Cloud computing is becoming mainstream for High Performance Computing (HPC) application development over the last few years. However, even though many vendors have rolled out their commercial cloud infrastructures, the service offerings are usually only best-effort based, without any performance guarantees. Cloud computing effectively saves the eScience developer the hassles of resource provisioning but utilization of these resources will be questionable if it can not meet the performance expectations of deployed applications. Furthermore, in order to make application design choices for a particular cloud offering, an eScience developer needs to understand the performance capabilities of the underlying cloud platform. Among all clouds, the emerging Azure cloud from Microsoft remains a challenge for HPC program development both due to lack of its support for traditional parallel programming support such as MPI and map-reduce and due to its evolving APIs. To aid the HPC developers, we present an open-source benchmark suite, Azure Bench, for Windows Azure cloud platform. We report comprehensive performance analysis of Azure cloud platform's storage services which are its primary artifacts for inter-processor coordination and communication. We also report on how much scalability Azure platform affords using up to 100 processors and point out various bottlenecks in parallel access of storage services. The paper also has pointers to overcome the steep learning curve for HPC application development over Azure. We also provide an open-source generic application framework that can be a starting point for application development for bag-of-task applications over Azure.
    Parallel and Distributed Processing Symposium Workshops & PhD Forum (IPDPSW), 2012 IEEE 26th International; 01/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: Cloud computing promises to offer great opportunities for research groups; however when researchers want to execute applications in cloud infrastructures many complex processes must be accomplished. In this paper we present the e-Clouds project which will allow researchers to easily execute many applications on public Infrastructure as a Service (IaaS) solutions. Designed for being a Software as a Service (SaaS) marketplace for scientific applications, e-Clouds allows researchers to submit jobs which are transparently executed on public IaaS platforms, such as Amazon Web Services (AWS). e-Clouds manages the on-demand provisioning and configuration of computing instances, storage, applications, schedulers, jobs, and data. The architectural design and how a first application has been supported on e-Clouds are presented. e-Clouds will allow researchers to easily share and execute applications in the cloud at low TCO (Total Cost of Ownership) and without the complexities associated with details of IT configurations and management. e-Clouds provides new opportunities for research groups with low or none budget for dedicated cluster or grid solutions, providing on-demand access to ready-to-use applications and accelerating the result generation of e-Science projects.
    Informatica (CLEI), 2012 XXXVIII Conferencia Latinoamericana En; 11/2012