Paper presents an advanced iterative MapReduce solution that employs Hadoop and MPI technologies. First, we present an overview of working implementations that make use of the same technologies. then we define an acad...
详细信息
Paper presents an advanced iterative MapReduce solution that employs Hadoop and MPI technologies. First, we present an overview of working implementations that make use of the same technologies. then we define an academic example of numeric problem with an emphasis on its computational features. the named definition is used to justify the proposed solution design.
Our research is focused on the simplification of parallel programming for distributed memory systems. Our goal is to build a unifying framework for creating, debugging, profiling, and verifying parallelapplications. ...
详细信息
ISBN:
(数字)9783642551956
ISBN:
(纸本)9783642551956
Our research is focused on the simplification of parallel programming for distributed memory systems. Our goal is to build a unifying framework for creating, debugging, profiling, and verifying parallelapplications. the result of this effort is an open source tool Kaira. In this paper, we focus on prototyping of parallelapplications. We have extended Kaira by the ability to generate parallel libraries. More precisely, we present a framework for fast prototyping of parallel numerical computations. We demonstrate our idea on a combination of parallel libraries generated by our tool Kaira and GNU Octave. Hence, a user can verify the idea in a short time, create a real running program and verify its performance and scalability.
When using Grid resources, usually a specific type of middleware is deployed. For example the European infrastructure is based on EMI-GLite and GLOBUS. the deployment of this middleware enforces a number of limitation...
详细信息
ISBN:
(纸本)9783662438800;9783662438794
When using Grid resources, usually a specific type of middleware is deployed. For example the European infrastructure is based on EMI-GLite and GLOBUS. the deployment of this middleware enforces a number of limitations concerning the OS, libraries and software packages. On the other hand the typical Grid middleware does not provide elasticity in the way Cloud services do. In our system we propose an entirely Java-based solution designed with elasticity in mind, providing distributed processing of queries from a single endpoint. the system is lightweight and easy to deploy on Cloud resourced and depends only on a few EMI packages. In this paper we describe the architecture and workflow of the system, providing benchmarks which show the effectiveness of our solution during processing of queries related to monitoring computational jobs in the Grid, using a distributed Cassandra database.
Modern genotyping technologies are able to obtain up to a few million genetic markers (such as SNPs) of an individual within a few minutes of time. Detecting epistasis, such as SNP-SNP interactions, in Genome-Wide Ass...
详细信息
the performance of Grid applications may be very unstable, especially when using workflows for job distribution. this is mainly due to the Grid overheads, like scheduling and queuing, introduced before the job is exec...
详细信息
ISBN:
(纸本)9783662438800;9783662438794
the performance of Grid applications may be very unstable, especially when using workflows for job distribution. this is mainly due to the Grid overheads, like scheduling and queuing, introduced before the job is executed on a worker node. Optimization problems using Genetic Algorithms (GAs) can be easily and efficiently implemented on Grids using Grid workflows. Due to the file dependencies introduced in the Grid workflows for GAs, mainly for genetic material interchange, these overheads are cumulative and thus very noticeable. this problem is also very evident when the jobs are short compared to the Grid overheads, i.e. the job spends more time waiting in a queue to be executed than the execution itself. In this paper we introduce a framework that enables users to easily utilize the Grid infrastructure for their optimization using GAs. It allows a user to preallocate certain number of pilot jobs, and also to dynamically manage their number for optimal availability of resources during the optimization process. In this way, once an application starts to execute the workloads, it will have at least one available pilot for execution of pooled tasks. this introduces better utilization of the Grid resources, as well boost the confidence in the infrastructure from users point of view.
We consider an emerging class of challenging networked applications called Real-Time Online Interactive applications (ROIA). ROIA are networked applications connecting a potentially very high number of users who inter...
We consider an emerging class of challenging networked applications called Real-Time Online Interactive applications (ROIA). ROIA are networked applications connecting a potentially very high number of users who interact withthe application and with each other in real time, i.e., a response to a user's action happens virtually immediately. Typical representatives of ROIA are multiplayer online computer games, advanced simulation-based e-learning and serious gaming. All these applications are characterized by high performance and QoS requirements, such as: short response times to user inputs (about 0.1-1.5 s); frequent state updates (up to 100 Hz); large and frequently changing numbers of users in a single application instance (up to tens of thousands simultaneous users).
In realizing futuristic services with agility, testbed environments that provide converged computing/networking resources based on SDN (Software-Defined Networking) and CC (Cloud computing) are becoming important. Fol...
详细信息
the Wigner equation provides an interesting mathematical limit, which recovers the constant field, ballistic Boltzmann equation. the peculiarities of a recently proposed Monte Carlo approach for solving the transient ...
详细信息
ISBN:
(纸本)9783662438800;9783662438794
the Wigner equation provides an interesting mathematical limit, which recovers the constant field, ballistic Boltzmann equation. the peculiarities of a recently proposed Monte Carlo approach for solving the transient Wigner problem, based on generation and annihilation of particles are summarized. the annihilation process can be implemented at consecutive time steps to improve the Monte Carlo resolution. We analyze theoretically and numerically this process applied to the simulation of important quantum phenomena, such as time-dependent tunneling of a wave packet through potential barriers.
parallel pattern libraries (e.g., Intel TBB) are popular and useful tools for developing applications in SMP environments at a higher level of abstraction. Such libraries execute user-provided code efficiently on shar...
详细信息
parallel pattern libraries (e.g., Intel TBB) are popular and useful tools for developing applications in SMP environments at a higher level of abstraction. Such libraries execute user-provided code efficiently on shared memory parallel architectures in accordance with well-defined execution patterns like parallel for-loops or pipelines. For heterogeneous architectures comprised of CPUs and accelerators, OpenCL has gained a lot of momentum. Since accelerated architectures do not provide a shared memory, it is not possible to directly use the approach taken in pattern libraries for SMP systems for OpenCL as well. In this paper, we are exploring issues and opportunities encountered by attempts to provide such patterns in the context of OpenCL. Based on a set of experiments with a scientific application on diverse OpenCL devices, we point out major pitfalls and insights, and outline directions for further efforts in developing pattern libraries for OpenCL.
暂无评论